Skip to main content

Full text of "The Philippine journal of science"

See other formats


The Philippine 
Journal of Science 

Volume 19 








Elmer D. Merrill, M.S., Editor 

R. C. McGregor, A.B., Associate Editor 

Albert H. Wells, A.B.; Granville A. Perkins, Ph.D.; A. P. West, Ph.D. 
T. Dar Juan, A.B., Phar.D.; F. Agcaoili, A.B.; A. S. Arguelles, B.S. 

Albert E. W. King 

Warren D. Smith, Ph.D.; Roy E. Dickerson, Ph.D. 

H. W. Wade, M.D.; Otto Schobl, M.D. 

F. G. Haughwout; Stanton Youngberg, D.V.M. 

Experimental Medicine 

LiBORio Gomez, M.D., Ph.D.; F. Calderon, B.A., L.M. 

Vicente de Jesus, M.D. 

Clinical Medicine 

W. H. Brown, Ph.D.; C. F. Baker, M.A.; H. Atherton Lee, M.S. 

0. A. Reinking, B.S.A., M.S.; L. M. Guerrero, Phar.D. 


Albert C. Herre, Ph.D.; C. F. Baker, M.A.; S. F. Light, M.A. 

C. S. Banks, M.A.; L. D. Wharton, M.A.; W. Schultze 


H. O. Beyer, M.A.; Otto Johns Scheerer, M.A. 

Anna B. Banyea, Copy Editor 


No. 1, July, 1921 

[Issued October 31. 1921.] 


Perkins, Granville A. The expression of the octet theory of va- 
lence in structural formulas 1 

One plate. * 

Light, S. F. Notes on Philippine termites, II 23 

Six plates and three text figures. 

Staff, O. A new species of Vincentia from the Philippines 65 

Wells, A. H. The preparation of tikitiki extract for the treatment 

of beriberi ^"^ 

RoHWER, S. A. The Philippine v^asps of the subfamilies Scoliinae 

and Elidinae '^^ 

Reinking, Otto A. Higher Basidiomycetes from the Philippines 

and their hosts, V ^^ 

McLean, F. T. The permeability of citrus leaves to water 115 

One text figure. 

Reviews ^^'^ 

No. 2, August, 1921 
[Issued December 9, 1921.] 

Lee, H. Atherton. Citrus-canker control: a progress report of 

experiments ^^^ 

Two plates. 

SCHEERER, Otto. Kalinga texts from the Balbalasang-Ginaang group 175 

One text figure. 

WiLEMAN, A. E. Notes on Japanese Lepidoptera and their larvae; 

Part VL... 209 

Two colored plates. 

Smith, Warren D. Ancient cave dwellers of Batwaan, Masbate, 

Philippine Islands 233 

Five plates and two text figures. 

Wharton, Lawrence D. Opisthorchis wardi, a new species of liver 

fluke from the cat in the Philippine Islands 243 

One plate. 

Light, S. F. Notes on Philippine Alcyonaria, Part VI: New Philip- 
pine Pennatularia (sea pens) of the genus Lituaria 247 



iv Contents 

No. 3, September, 1921 

[Issued December 19, 1921.] 


Maxwell, J. Preston. Filariasis in China 257 

Twenty-five plates and four text figures. 

Merrill, Elmer D. A review of the new species of plants proposed 

by N. L. Burman in his Flora Indica 329 

No. 4, October, 1921 

[Issued January 16, 1922.] 

Reinking, O. a., and Groff, G. W. The Kao Pan seedless Siamese 

pummelo and its culture 389 

Sixteen plates and one text figure. 

Gebien, Hans. Philippine Tenebrionidae, II - 439 

Two plates. 

Banks, Charles S. A Philippine nemestrinid (Diptera) 517 

One plate. 

No. 5, November, 1921 
[Issued February 15, 1922.] 

Heller, K. M. New Philippine Coleoptera 523 

Three plates. 

No. 6, December, 1921 
[Issued March 20, 1922.] 

•Lantin, Pedro T. Various methods of serum application in bacillary 

dysentery 629 

Six text figures. 

Perkins, Granville A. Unsymmetrical addition to the double bond, 

I: A theory of the reaction mechanism of the direct union 645 

Six text figures. 

Groves, James. Charophyta from Annam and Guam... ... 663 

ROHWER, S. A. The Philippine wasps of the subfamily Sphecinae.... 665 

Merrill, Elmer D. Two new species of plants from Hainan 677 

Cole, Howard Irving. The dissociation of hexaphenylethane from 

the viewpoint of the octet theory of valence 681 

McGregor, Richard C. New or noteworthy Philippine birds, IV.... 691 

Four plates and three text figures. 

Becker, Wilhelm. Die Violen der Philippinen 707 

Okamoto, Kikuo. Secondary sexual characters in the loach Misgur- 

nus anguillicaudatus Cantor 723 

One plate. 

Perkins, Granville A. The structure of chlorine dioxide and relat- 
ed compounds 729 

Contents y 


Welles, Colin G. Cercospora leaf spot of coffee 741 

One plate. 

Welles, Colin G. Cercospora leaf spot of Averrhoa carambola 747 

Two plates. 

Review 753 

Errata 755 

Index 757 

Vol. 19, No. 1 

July, 1921 

The Philippine 
Journal of Science 






Published by the Bureau of Science of the Government of the Philippine IsIjidgU' 

Elmeb D. Merrill, M.S., Editor 

R. C. McGregor, A.B., Associate Editor 

Albert H. Wells, A.B.; Granville A. Perkins, Ph.D.; A. P. West, Ph.D. 
T. Dar Juan, A.B., Phar.D.; F. Agcaoili, A.B.; A. S. Argubllbs, B.S. 

Albert E. W. King 


Warren D. Smith, Ph.D.; Roy E. Dickerson, Ph.D. 


H. W. Wade, M.D.; Otto SchSbl, M.D. 

F. G, Haughwout; Stanton Youngberg, D.V.M, 

Experimental Medicine 

Liborio Gomez, M.D., Ph.D.; F. Caldbron, BA., L.M. 

Vicente de Jesus, M.D. 

Clinical Medicine 

W. H. Brown, Ph.D.; C. F. Baker, M.A.; Atherton Lee, M.S. 

0. A. Reinking, B.S.A., M.S.; L. M. Guerrero, Phar.D. 


Albert C. Herre, Ph.D.; C. F. Baker, M.A.; S. F. Light, M.A. 

C. S. Banks, M.A.; L. D. Wharton, M.A.; W. Schultzb 


H. 0. Beyer, M.A.; Otto Johns Scheerer, M.A. 

Anna B. Banyea, Copy Editor 

Manuscript intended for publication and books for review should be sent 
to the editor. One hundred separates of papers published in the Journal 
are furnished to authors without charge. Additional copies may be had at 
authors' expense. 

Publications sent in exchange for the Philippine Journal of Science 
should be addressed: Library, Bureau of Science, Manila, P. I. 

The Journal is issued twelve times a year. The subscription price is 6 
dollars. United States currency, per year. Single numbers, 50 cents each. 
' Subscriptions may be sent to the Business Manager, Philippine Jour- 
nal of Science, Bureau of Science, Manila, P. I., or to any of the agents 
listed on the third page of this cover. 

The Philippine 
Journal of Science 

VOL. 19 JULY, 1921 No. 1 


By Granville A. Perkins 
Chemist f Bureau of Science , Manila 

The science of organic chemistry, as we know it to-day, may 
be said to owe its very existence to the idea of structural formu- 
las developed by Kekule, Frankland, and Couper, sixty years ago. 
In the last two decades, however, the development of both organic 
and inorganic chemistry has been greatly retarded by the fact 
that Kekule's simple ''affinity units'' do not represent with 
sufficient accuracy the actual forces which bind atoms together. 

Recent attempts of physicists to apply the new knowledge of 
electrons to the fundamental problem of chemistry, namely, the 
nature of chemical affinity, met with only partial success, until 
Langmuir ^ finally showed that certain recent ideas of atomic 
structure, notably those of Lewis, ^ can be used to form a re- 
markably successful working hypothesis in both organic and in- 
organic chemistry. The conception of electron shells and shared 
electrons as presented by him gives one a definite picture which 
is undoubtedly very close to the actual nature of the union 
between atoms. 

Langmuir's ''octet theory of valence" is so simple and exact 
that it will be found of great value to chemists, from the over- 
curious student who asks his professor what makes the atoms 
stick together to the investigator who wishes to predict the 

' Received for publication, February 28, 1921. 

' Langmuir, I., Journ. Am. Chem. Soc. 41 (1919) 868, 1543; 42 (1920) 274. 

'Lewis, G. N., Jaurn. Am. Chem. Soc. 38 (1916) 762. 

180365 1 

2 The Philippine Journal of Science lo-^i 

results of a reaction which has never been performed. The 
method used by Langmuir in applying the theory, however, iap- 
pears to the writer to be somewhat cumbersome for general use, 
and to obscure, by its indirectness, some of the value of the 
theory. The purpose of the present article is to present a sys- 
tem for writing structural formulas which will be as simple 
and direct as possible, and at the same time represent the mole- 
cules as accurately as possible in terms of modern atomic theory. 
It is hoped not only to furnish by this means a simple method 
for practical application of the theory in its present stage of 
advancement but even to develop certain valence relationships 
which have not hitherto been clearly expressed. 

While it is assumed that anyone interested in the subject 
is already familiar with recent developments of the octet theory 
in the hands of Langmuir and others, for the sake of clarity the 
subject will be briefly reviewed. 


In the light of recent physical evidence the essential portion 
of any atom is a minute nucleus composed of positive units of 
electricity (sometimes called positive electrons) and a smaller 
number of negative units (sometimes called negative electrons, 
but usually simply electrons) very closely packed together and 
bound by the most powerful forces known. The positive units 
are all identical, each having a mass of approximately 1, ex- 
pressed in atomic weight units. Similarly the (negative) elec- 
trons are all identical, but have negligible mass. Each electron 
in the nucleus neutralizes one positive unit, so the total outside 
electrical effect, called the nuclear charge, is measured by the 
number of positive units less the number of electrons (negative 
units) . This difference is called the atomic number of the nu- 


Except under very unusual circumstances, such as when trav- 
eling with extreme velocity, a nucleus is never found alone, be- 
cause it normally attracts as many electrons as its atomic num- 
ber, forming an electrically neutral atom. These electrons do not 
enter the nucleus but arrange themselves in nearly spherical 
concentric shells, which vary in number from one to seven ac- 

*Harkins, W. D., Jonrn. Am. Chem. Soc. 42 (1920) 1956; Rutherford, 
E. E., Proc. Roy. Soc. London 97A (1920) 374-401. 

19, 1 Perkins: The Octet Theory of Valence 3 

cording to the number of electrons. The shells are filled in the 
following order: First shell, 2 electrons; second, 8; third, 8; 
fourth, 18; fifth, 18; sixth, 32; seventh, (32?). (See Table 1). 
The electrons either rotate, perhaps in ring form, or revolve in 
small '* orbits, so that they are pov^erful electro-magnets. (See 
Plate 1.) 

TABLE 1. — The atoms arranged according to their electron shells. 

I Hydro- 
1 gen 

Number of shells-.. 
Electrons in kernel 
Completed shell — 


(*0) (He 2) 

(HI) I Li 3 

! Be 4 








(A 18) 
K 19 
Or 24 


Mo 42 

He 2 


Si 14 
CI 17 

Co 27 
As 33 
Br 35 



In 49 


La 57 
* 61 
Ho 66 
Dy 67 

Tm2 70 

* 75 
08 76 









• 85 


* 87 


Ac 89 


Ux2 91 




trons in 



















Vacan- i 
cies in ' 

* Not yet discovered. 

= Concerning the size of these rings or orbits, there is much difference of 
opinion among physicists. They may be much smaUer, relative to the shell, 
than represented in Plate 1, but larger orbits than these seem to be excluded 
by chemical evidence. 

4 The Philippine Journal of Science 1921 


The completed inner shells of an atom (that is, all the shells 
except the outermost) together with its nucleus, constitute 
the kernel. This is bound together by such strong electrostatic 
and magnetic forces that it is never disrupted in ordinary chem- 
ical reactions, but acts as an unchangeable unit except under 
extremely penetrating forces, such as those of X-rays. 


For brevity the outer shell is called simply the shell. Unless 
this is complete, as in the helium group, the neutral atom exerts 
•forces, both magnetic and electrostatic, effective at a considerable 
distance. It is upon these forces that all chemical action depends. 

This is, roughly, the theory of atomic structure which seems 
to meet with general acceptance among chemists at the present 


Two types of chemical union have been distinguished ; namely, 
primary valence unions and secondary valence unions. The 
present paper deals chiefly with the primary type, which involves 
the main uniting forces of atoms. This has been subdivided into 
two kinds, which may be called salt-forming unions and direct 

Salt-forming unions are caused by the fact that an atom with 
a nearly complete shell has a strong tendency toward completing 
its shell, and is able to appropriate from another atom with a less 
complete shell one or more electrons. Both atoms thus become 
charged and attract each other. This type of union has often 
been represented as follows: 

Na"^ CI 

Direct unions have been a much more puzzling problem, but 
Lewis ^ has advanced the explanation, later developed by Lang- 
muir, that they are brought about by the sharing of one or more 
pairs of electrons by two atomic shells. For such a union to take 
place it is necessary that both atoms have a tendency to complete 
their shells. Since, in such a case, neither can detach an electron 
from the other, they compromise by sharing 2, 4, or 6 (but not 1, 

« Lewis, G. N., loc. cit.; cf. Bohr. N., Phil. Mag. 26 (1913) 1476; Parson, 
A. L., Smithsonian Misc. Coll. 65, No. 11; Milliken, R. A., Science 45 
(1917) 321; Silberstein, L., Phil. Mag. 39 (1920) 46. 

^ Loc. cit. ; cf . Parson's "group of fourteen," op. cit., p. 29. 

19,1 Pey^kins: The Octet Theory of Valence 5 

3, or 5)^ electrons, which then do duty in both shells. Thus two 
chlorine atoms, having incomplete shells of 7 each, can form two 
complete octets in which one pair of electrons is shared. This 
kind of union, which will be called in this paper a direct union, 
was represented graphically by Lewis as follows: 

: c\ : CI : 

Langmuir has simplified the formulas, using a single line to 
represent a shared pair : CI— CI, HO— CI— 0, 0=N— 0— N=0, 
N=N=0, etc. He has shown that a large number of hitherto 
perplexing structures can be readily expressed in this way. 


Chemists have always tried to make generalizations concern- 
ing the valence (that is, the quantitative combining power) of 
atoms. An inspection of the Langmuir formulas above tells one 
nothing of the true valence relations of the atoms. He calls the 
number of lines attached to any atom (that is, the number of 
pairs which it shares in that compound) its covalence, and has 
developed some valuable generalizations regarding this variable 
property. What have been considered the true valence relations, 
however, he consigns to the equation : ^ 

e zrzz 8n — 2p. 

It seems to the writer that there are two different kinds of 
direct union not hitherto distinguished. By recognition of this 
difference it is possible to construct formulas of considerable 
graphic value, at the same time dispensing with the equation. 

The proposed system is based on positive and negative valence, 
the maximum values of which are clearly expressed by 
Langmuir ^^ as follows : 

Now the maximum positive valence is a definite conception — it repre- 
sents the total number of available electrons in the shell * * * On the 

' In the case of benzene and similar compounds, however, the writer 
believes that 3 electrons are shared. 

* This equation expresses the fact that in a molecule the total number 
of outer shell electrons (e) exist in completed octets, which requires 8 
electrons per octet (871) less 2 for every pair shared ( — 2p). The inter- 
pretation of the letters is somewhat modified for the case of hydrogen, 
where the completed shell has 2 electrons, and the heavy atoms, whose 
completed shells have 18 or 32 instead of 8 electrons. 

'•Langmuir, I., Journ. Am. Chem. Soc. 41 (1919) 926. 

6 The Philippine Journal of Science 1921 

other hand, the maximum negative valence represents the number of 
electrons which the atom must take up to reach a stable, form like that of the 
inert gases. 

Langmuir further brings out the facts, at least by inference, 
that the actual negative valence in any compound, if exhibited at 
all, is almost always the same as the maximum, but the actual 
positive valence is often less than the maximum, giving two or 
more classes of compounds of the same metal, such as the cu- 
prous and cupric compounds. 

It seems important to the writer to express graphically, when 
writing structural formulas, the actual positive or negative va- 
lence exhibited by each atom. 


1. Valence, — Represent the maximum electronegative valence 
of an atom in the ordinary way: 

H— , F— , -0-, -N-, C , B-. 

I I / \ 

Each line represents the organic chemist's '^unsatisfied bond," 
the physical interpretation of which is vacancy for one more 
electron in the shell. The nitrogen atom, for example, needs 
three electrons to complete a shell of eight. The number of 
''unsatisfied bonds'' for each electronegative atom can readily 
be found by consulting Table 1. Electropositive valence is on 
no account to be represented by "unsatisfied bonds" as it never 
causes direct union between atoms. It may be represented as 
follows : 

NaS CaI^ AH". 

2. The salt-forming union, — It is evident that an atom may fill 
a vacancy in its shell (satisfy a bond) by simply acquiring an 
electron, thus becoming a negative ion.^^ It may even create 
more bonds by the reverse process: F — becomes F , — ^0 — be- 

I + 

comes , H — becomes — H — , — N — becomes ~ N — . 

I I 

Negative ions having no unsatisfied bonds form stable saltlike 
compounds with metallic ions: 

Na-^ Cr, Ca+^ zzzzzz 

It is to be noted that while the formation of ions is due to the 
shell-completing forces of one of the atoms involved, the union 

** For the lack of a better w^ord, this term seems to be quite generally used 
for a charged atom even though it may not be mobile. 

19,1 Perkins: The Octet Theory of Valence 7 

between the ions is due to electrostatic attraction. This is the 
salt-forming union, and will be represented by a dotted line be- 
tween the ions. 

3. The normal direct union. — When two atoms are held to- 
gether due to the fact that the shell-completing forces of both 
atoms act on a pair of electrons which is shared between them, 
the union may be said to be direct. Such a union is to be repre- 
sented by the usual line for a "satisfied bond" if it is normal; 
that is, if one electron of the shared pair has been supplied by 
each atom. In this case it is evident that each shell involved has 
filled one electron vacancy by the process of sharing and has thus 
"satisfied'' one "bond.'' 

Examples: H— H, H— 0— H. 

4. The borrowing direct union, — A direct union in which one 
atom supplies both electrons of the shared pair may be called 
a borrowing union. In this case the borrowing atom fills two 
vacancies in its shell (that is, satisfies two bonds), and the 
lending atom neither gains nor loses electrons. A convenient 
way of representing such a union, whereby — A — satisfies two 
bonds, and B none, is A 3o B. The sign x) has here neither its 
mathematical nor its astronomical significance, but has consider- 
able graphic value in representing that two "valence bonds" of A 
and none of B, are satisfied. 

Examples: H - Cl« CI — P ex 


5. Double and triple bonds. — Each symbol for a normal union, 
— , or for a borrowing union, 20^ represents one shared pair of 
electrons. If two or three pairs are shared by two atoms each 
part of the total union is represented by the appropriate symbol. 

Examples: N ^ N - 0, H - C C - H. 


An atom which shows a tendency to become positively charged 
is called an electropositive atom (in the chemical sense). No 
atom actually repels one of its electrons to a very long distance, 
so this electropositive tendency is really only a comparative 
weakness which some atoms have in holding their electrons 
against any outside detaching force. In electropositive atoms 
the forces binding shell electrons seem to be calculable as ordi- 
nary electrostatic forces.' '" Therefore the smaller the kernel, 

'' Langmuir, I., Journ. Am. Chem. Soc. 41 (1919) 877. 


The Philippine Journal of Science 


the greater the retaining force ; and, in the case of two or more 
removable electrons, each succeeding electron is much more diffi- 
cult to remove. Accordingly Cs i is the most electropositive 
atom, and it is seldom that we find more than two or three elec- 
trons completely removed from any atom, no matter how many 
shell electrons it may have. Four, five, six, seven, or eight elec- 
trons may be partially removed, however, passing into the ^ells 
of other atoms, which remain closely bound to the atom in ques- 

Gradations in the closeness of the salt-forming union due to 
these differences in the electropositive tendency may be ex- 
pressed roughly by varying the length of the dotted line. 







Ti :----- 



Due to the closeness of union in the oxides of high valence 
they do not form basic hydroxides, involving rupture of the 
metal-oxygen union by ionization, but the central atomic kernel 
attracts, by its strong electrostatic field, even an excess of oxygen 
ions, resulting in the formation of a negatively charged aggre- 
gate as: 

Ti : 


0::: Cr -- - Cr 

19.1 Perkins: The Octet Theory of Valence 9 

It is obvious that in any aggregate of an atomic kernel and 
20 , 30 , 40 and in such complexes as Cr^O^, an octet 
way be formed around the kernel, and in many cases probably is 
formed. Therefore the formulas of such aggregates in which the 
central atom has a valence of 4, 5, 6, or 7 could be written in ex- 
actly the same manner as those of the analogous ions containing 

— C — , — N — , — S — , and CI — , which also contain 4, 5, 6, 

I I 

and 7 electrons in the shell, respectively. The distinction, as 

it appears to the writer, is that in one class of cases all the 
known facts are as well explained on the basis of ordinary 
electrostatic attraction alone (subsequent to ion formation) as 
they are by assuming an effective tendency of the electrons sur- 
rounding the kernel in question toward octet formation, but 
that in another class of cases we have definite evidence that 
there is an effective tendency toward octet (or other stable shell) 
formation. The writer prefers to use the salt-forming symbol 
for the former class ^nd the direct symbol for the latter. 


An atom is called electronegative (in the chemical sense) 
when it shoWs a tendency to become negatively charged. As 
has just been intimated, the distinguishing characteristic of 
a negative atom is that it shows a definite tendency toward 
building up some stable arrangement, usually an octet, of elec- 
trons. A discussion of the forces involved is beyond the scope 
of this paper. ^^ It is necessary only to point out that there are 
observed differences in the electronegative tendencies of atoms. 
Fluorine is the most strongly negative, and starting from this 

** It is illogical to try to apply very closely to this case the laws of elec- 
trostatics as we know them. Latimer and Rodebush, Journ. Am. Chem. Soc. 
42 (1920) 1425, treat electronegativity practically as an ordinary elec- 
trostatic phenomenon, arriving at the conclusiotn that "In one sense then, 
hydrogen is the most electronegative of all the elements.'' Langmuir, 
Journ. Am. Chem. Soc. 41 (1919) 908, approaches this problem more 
reasonably, concluding that "Hydrogen therefore can hardly be classed 
as an electronegative element.'' He proceeds, however, to apply (page 
910) the inverse square law to the total force between the nucleus 
and the shell in the case of carbon and other atoms, although later 
(page 932) he suggests the fact that the whole existence of th© shells 
depends V on some such balance of forces as a discontinuous inverse square 
attraction opposed by an inverse cube repulsion. 

It would seem necessary to use caution in applying any force laws to 
the shell electrons, especially regarding the attraction of the nucleus. 

10 The Philippine Journal of Science 1921 

atom in Table 1, one finds a graded weakness in going either to 
the right or upward. Another observed fact is that the electro- 
negative tendency of an atom varies according to the atoms com- 
bined with it, in a manner which shows that electrons in a shell 
are held by forces of an elastic nature, and that they shift their 
positions of equilibrium under the influence of outside electro- 
static forces. 

Due to these natural and acquired differences in electro- 
negative powers, it is only in such symmetrical cases as 
CI — CI, H3C — CH., that a shared pair of electrons is shared 
equally by two atoms. If A is more electronegative than B, 
in general the shared pair will be held more closely by A than 
by B, in such a manner that A will become negatively charged 
as compared to B, thus: A~ — B ^. Such a union has long been 
called a polar union, and is generally represented by an arrow 
indicating a partial electron transfer: A<-B. There is no 
difficulty in introducing the arrow into the proposed system 
in cases where it is desired to point out polarity, as H-^Cl, 
H^O<-H. As the complete polarity of the salt-forming union 

has already been well represented by the sign ~ "^, the arrow 

will be used only to denote polarity in direct unions. 

The distinction and relation between a borrowing union and 
a polar union should be clearly understood. The borrowing 
union sign Ax)B indicates that both electrons of the shared 
pair belonged exclusively to the lending atom, B, before the 
union took place. The polar union sign A^-B has in the past 
signified nothing as to the origin of the shared pair but only 
that after union there is an electrostatic dipole A<— B+. 

There are then two cases of polar union, depending on 
whether the union is normal or borrowing. From the defi- 
nition it is evident that the borrowing union A ^ B is essentially 
a polar union, because the borrowing atom — A — acquires an 
interest in two electrons with which it originally had no 
connection. Therefore if — A — was originally neutral, it 
becomes a negative pole. In fact, Ac>o B can be written A~— +B. 
Nevertheless, A shows no tendency to leave B in the form of 
a negative ion, and the shared pair is almost always held more 
closely by B than by A ; so the distinct symbol A 30 B should 
be retained, and the symbol A^B used only to represent a 
normal polar union; that is, one in which each atom supplies 
one electron of the shared pair, and in which after union the 
shared pair is more closely associated with A than with B, 
so that an electrostatic dipole is formed. The union A^B, 

19,1 Perkins: The Octet Theory of Valence H 

when ruptured, practically always gives the products — A — 
and B. The union A<^B may give A— and — B, but usually 

A and ^ b"" . 

Sometimes it is convenient for comparative purposes to 
express a loose, normal union by a long connecting line, and 
a close, strong union by a short line, as was done with salt- 
forming unions. A loose union is not necessarily polar: 

F — F I I. 

All of these distinctions of gradation are entirely unnecessary 
in an ordinary structural formula representing merely the out- 
standing valence relations, but are very valuable when attention 
is to be called to certain comparative characteristics. 


The mode of application of the proposed system to all of the 
atoms will be briefly indicated. It is believed thati the rela- 
tions of the chemical properties of the atoms to the present 
theory of atomic structure will best be seen by grouping them 
(except H — and the group Sa — Lu) according to the horizontal 
lines in Table 1. 


H— . 

Electrons in shell = 1. 
Vacancies in shell = 1. 

The hydrogen atom is unique in that its bare kernel (in this 
case the nucleus) can acquire a complete shell simply by 
attaching itself to any convenient pair of electrons not already 
shared by two kernels. This property accounts for the peculiar 

mobility of the kernel, — H — , in molecules, which led to per- 
plexing controversies between organic chemists until it was 
finally recognized and called ''tautomerism." This property 
also distinguishes H — radically from all the other atoms 
having a negative valence of 1, giving it a pseudo-electropositive 
character. As Latimer and Rodebush '* point out, the ionization 

of acids is not direct, like that of salts, but depends on a 

mobile (tautomeric) union of the — H — with molecules of the 

solvent. Thus H — CI may be considered to give — H [- CI , 

'^Latimer, W. M., and Rodebush, W. H., Journ. Am. Chem. Soc. 42 
(1920) 1425. 


The Philippine Journal of Science 


but only because the unsatisfied bonds of — H — are immediately 
satisfied by the formation of a solvated ion, such as H -- - - H. 


This hydrated hydrogen ion may also be written : 


















The advantage of formula (2) is that it shows the symmetry 
of the compound. It is readily seen that a borrowing bond 
may be considered as a normal bond preceded by the transfer 

of an electron. Imagine —0 H H— to become by transfer 

— H, which then combine by normal union. Ac- 

tually the transfer does not take place first, but at the moment of 
union there is a distribution of the positive charge. This fact 
is best expressed by formula (4), but (2) is a simpler 
expression of valence relations. 


Na' H, H — 0-H, H 



He", Ne^ A^ Kr^ Xe^ Nt^ 
Electrons in shell = 0. 

These atoms have zero valence, but according to the octet 
theory they may possibly have the power of combining with 

oxygen or with — H — as follows: 

Xe -i 0— giving Xe^c 0, Xe 4- -H— giving Xe o: H . 

If so, the combination would be very unstable, as these 
' inert'* atoms have very weak external fields. 

19, 1 Perkins: The Octet Theory of Valence 13 

I I I I I 

Li , Na , K , Rb , Cs . 
Electrons in shell = 1. 

The atoms of this group have an electropositive valence of 1. 
Their outer shells are too incomplete for the formation of 
completed shells, and therefore they enter only into salt-forming 
unions : 

Li' CI , Li^ ~ Li', 

Li"^ H, Na"^ CI, Cs^ CP. 


Be , Mg , Ca , Sr , Ba , Ra . 
Electrons in shell = 2. 

These atoms are similar to the lithium group except that 
each loses 2 electrons: 

Be zzzzzzO , Cr Ca^"^ CF, 

Br Ra^^ Br~ 



B ( - B - ), Al ( - Al - ), Sc , Yt , 

/ \ / \ 


La , Ac . 

Electrons in shell == 3. 

This group is predominantly electropositive. Under favor- 
able circumstances, however, the first two members can complete 
their outer shells, thus exhibiting a negative valence of 5. That 
this property stops abruptly with AP" is due to the fact that 
Sc^" would require 15 electrons, instead of 5, to complete its 

The compounds in which B is electropositive are not at all 
typical saltlike compounds. The negative ions surround the 
small B+++ kernel so completely and so closely that the exter- 
nal field is small and rupture of the union very difficult." 

" cf . Langmuir, ibid. p. 929. 


The Philippine Journal of Science 






Na^ B 



H - 0- B — 0- H 






- Al 



F - H 


Ti , Zr , Ce , Th . 


V , Cb , Pr , Ux2 . 

(II, III) VI (III, IV) VI III (IV?) (in. IV, V) VI 

Cr , Mo , Nd , U 




(11) in 


ni (IV?) 


Sa , Eu , G(J ,' Tb ' , Ho , 


Dy , Er , Tm , Tma , Yb , Lu 

(II. IV) V 

(II. IV. V) VI 



(II) HI (IV) II. Ill (IV, VI, VII. VIII) II. in. IV, VI, VIII 

Fe , Ru ,0s i 

n (ni) (ID III (IV) n. in. iv 

Co , Rh , Ir 

n (III) II. (in. IV) II, IV (VI) 

Ni , P(3 , Pt 

I. n I I. Ill 

Cu , Ag , Au 

II (I?) II I. II 

Zn , Cd , Hg . 

(II) III (I, II) HI I, in 

Ga , In , Tl . 

Electrons in shell = 4 to 27. 

19, 1 

Perkins: The Octet Theory of Valence 


From what has been said one would expect the common va- 
lence of these atoms to be about 3, and the maximum to be 8 in 
Sa and all atoms below it in Table 1. The actual valences, how- 
ever, have been successfully explained by Langmuir^^^ on the 
basis of the stability of certain partly completed shells. The 
shells of Ni, Pd, Er, and Pt, can have a stability remotely resem- 
bling that of the inert atom shells, but only when rearranged in a 
form (the p form) not stable except when surrounding a kernel 
more highly charged than the kernels of these respective atoms. 
Therefore, some of the atoms somewhat below Ni, Pd, Er, and 
Pt in Table 1 tend to lose electrons until they have a pseudo- 
kernel of the form p-Ni, p-Fd, i8-Er, and p-Ft. This is made 
possible only by the rearrangement just mentioned, and there- 
fore does not affect any atoms above these in the table. 

There are probably slight electronegative tendencies in some 
of this large number of atoms, that is, forces tending toward 
completion of certain stable arrangement of shell electrons. 
Any such forces are so weak, however, that we have no evidence 
of them except, perhaps, in a few compounds like NajZrFy, 
[Co (NH3) ,C1] Cl„ [Pt (NH3) ,C1 J CI,. 

It is usual to make somewhat larger groups of the atoms, but it 
seems to the writer that the grouping in Table 1 shows most 
plainly the relations between the structure of atoms and their 
chemical properties. The partial relation between such groups 
as the chromium and sulphur groups has already been pointed 
out, and it is easily seen that the partial resemblance of p-Ni, 
p-Pdy p-Er, P'Ft to the inert atoms causes a number of partial 
similarities, such as those between the copper and lithium groups 
and the zinc and beryllium groups. 


Ti :-: 0~ 

Cr "i: 

Ti --: 



0-- ^+ 

:; + 

0""^ J 

( i + 


• 1 + 


11 -1- 

Cr :i:iO"~ 

1 1 + 

1 1 


Os ::i:0 

1 1 

1 1 
1 1 

1 1 



"Ibid. p. 876. 

16 The Philippine Journal of Science. 1921 

ci Ci 

HsN I NH3 > 

Pt -:' 

H3N I NHg'^^---^, 
CI ^' 


i I II n I 

-C-, ~Si- -Ge— (Ge ), Sn (-Sn-), 

111 I 

(I) II (HI) I 

Pb (— Pb - ). 


Vacancies in shell = 4. 

There is unmistakable evidence that there are definitely four 
vacancies in the shell of each of these atoms. Sn" and Pb" 
are predominantly electropositive; but, as the relation of these 
atoms to the ''inert group" has been established by physical evi- 
dence, each containing 4 less electrons than the corresponding 
inert atom, we have good reason to believe that this fact plays 
a large part in determining their chemical properties. This idea 

is substantiated by the behavior of — Sn — and — Pb 


organic compounds. 

The ordinary carbon-hydrogen-oxygen compounds are ex- 
pressed in the proposed system exactly in the same manner as is 
customary among organic chemists. The peculiar compound CO, 
however, has never been successfully explained except by Lang- 
muir's hypothesis that the two kernels share 5 pairs of electrons, 
1 pair being held rather closely by the kernels and the other 4 
pairs in an octet external to both kernels. In this case the va- 
lence of each atom is increased by 2, because a stable arrange- 
ment of 10, instead of 8, shell electrons is formed- 


c'^O = C-0 Pb^-:-::0 

= Pb = 


-1+ 1 



Sn ----_-r ^}- H - Ge — H 




19.1 Perkins: The Octet Theory of Valence 17 


I li \ 

H - C - - C - - H - 





II I I in (ID III 

- N -, - P -, - As -, -Sb - (Sb ), Bi 

( - Bi - ). 

Vacancies in shell =3. 

The system of structural formulas used by organic chemists 
has never been applied successfully to compounds containing 
'*pentavalent" nitrogen. According to the proposed system ni- 
trogen never shows in these compounds a valence above 3, al- 
though, as pointed out by Langmuir, the covalence is 4. The 
three cases where nitrogen has a peculiar valence of 5 are 

(:_) (■ ) (££) 

N ^ N, N =^ 0, and — C N and are explained by Langmuir as 

being "isosteric'' with CcoO. 

The cases where nitrogen has been supposed to show a high 
valence are explained either by the borrowing union or by the 
formation of a positive ion. The formation of ammonium salts 
is strictly analogous to the reaction between H — CI and 
H — — H, described in connection with hydrogen. The ammo- 
nium ion, however, is much more stable than OH^ 


I + 
|H -- N - H 


I H 

In the case of nitrogen oxides and oxy-acids N has such a 
tendency to a covalence of 4, that it lends electrons, especially 
to — — , as : 

'^ DO N = 

=N -N- 

- II II 
= 0»N-NccO 

0=oN - 

180366 2 

-0 -NccO 



18 The Philippine Journal of Science 1921 

It would not be expected, however, that — N — could borrow 
from a strongly negative atom as — — , forming such a com- 
pound as N 5o oc N. 

In a few compounds the heavier atoms of this group show elec- 
tropositive tendencies. The salts of Bi "i are good examples 
of how the weak forces toward completion of a shell may be 
entirely overcome by some stronger shell, and electrons lost, 
rather than gained, by the weakly negative atom. It has often 
been considered that in the formation of pentahalides the atoms 
of this group show a positive valence of 5. A comparison with 
vanadium, which of all the positive atoms has the greatest simi- 
larity to the trivalent negative atoms, is of interest. The most 
evident difference is that the halides and oxyhalides of this 
group indicate valences of 3 and 5, but never 4. Vanadium, as 
would be expected from the consideration of electrostatic forces, 
shows also a valence of 4. 

It is evident then, that the trihalides of this group have a 
stability which does not permit of the addition of one more halo- 
gen atom. When two are added they probably form a compound 
strictly analogous in structure to NH^CI in most, if not all, 
cases.^"^ Thus, PCI3 is tetrachlorophosphonium chloride: 


\ / 

/ \ 

Cl Cl 


_0— — S— , — Se— — Te— , Ra F. 
Vacancies in shell = 2. 

In most compounds containing borrowing unions it is found 
that either — — , or — S — is the borrowing atom. This is 
not surprising, as of all the atoms capable of acting in this way 
these two are the most electronegative. It is the borrowing 
union which explains the fact that the number of oxygen atoms 
which can attach themselves to a negative atom depends more 
on the size of that atom than on its valence. Thus we have 
ClOr, SOr ~, POr-- ~ Si04 , but N03~ and CO3 ". 

The sulphonium and oxonium salts, so familiar to organig 
chemists, are assigned formulas analogous to those of ammo- 
nium salts. 

" cf. Langmuir, ibid, p. 919. 

19, 1 Perkins: The Octet Theory of Valence 19 

In comparing the halides of this group with those of Cr, Mo, 
W, and U, we find even stronger indications of definite shell- 
completing action than were noted in the preceding group. 
There seems to be no other explanation for the avoidance by 
S, Se, and Te of the apparent valences 3 and 5 except that they 
retain always their negative valence of 2, attracting halogen 
atoms beyond this amount only in pairs, and in the same way 
that NH3 combines with HCl. 



OcoS=0 8 






H3C - S -- 

F F 


-I F---S— -F 


-S --C1 


/ \ 
F F 


F— , CI—, Br—, 
Vacancies in shell = 





As these are typical negative atoms, there is no difficulty in 
representing their compounds. 

F — F F— Br — F I ++. 


I I ~~-F 

F F 


I t-.d 

H— 0—10=0 


.. 1 + 








1 + 



.. I 




20 The Philippine Journal of Science 


Anyone who has found it necessary to use as guides in ex- 
perimental work such valuable but hazy and incomplete valence 
theories as those of Werner, Thiele, Friend, and Nef will wel- 
come Langmuir's octet theory of valence as the true ''key to the 
situation." It is hoped that the proposed system of structural 
formulas will be found valuable not only as a method for rep- 
resenting the primary valence relations of atoms in molecules 
but as a better basis than has been available in the past for the 
study of polarity, secondary valence, reactivity, selective absorp- 
tion, and other phenomena depending on the shell electrons. It 
has enabled the writer to see certain perplexing reactions in a 
new light, and has led to the formulation of a theory of reaction 
mechanism of the direct union, which will be published in the 
near future. 


1. A system for writing structural formulas has been devised, 
based on the octet theory of valence as presented by Langmuir. 

2. The new feature of the system is a distinction between 
''normal" and "borrowing" unions which enables the actual 
valence relations of the atoms to be represented in the formulas. 

3. The borrowing union, A^oB, is unique in that it is polar in 
the direction A — "^B although the shared pair is held more 
closely by B. 

4. The mode of application of the system to all of the known 
elements has been indicated. 

5. The system in itself represents only the primary valence 
relations of the atoms, but can easily be adapted to the study 
of other phenomena depending on the valence electrons. 


Plate 1. Models of the Electron Shells 

As the exact forces (see p. 3) acting on shell electrons are not known, 
these models are to be considered only very rough approximations to the 
actual proportions and arrangements. The arrangements shown are based 
on the assumption that magnetic attraction is the determining force. The 
white rings represent electrons revolving in orbits, or in actual ring shape, 
in a direction clockwise* to the observer, the black ones, counter-clockwise. 
Paper disks are placed inside to increase the visibility. 









By S. F. Light 
Professor of Zoology^ College of Liberal Arts, University of the Philippines 


This paper presents descriptions of six species of Philippine 
termites which seem to be new to science. They represent four 
genera {Kalotermes, Cryptotermes, Prorhinotermes, and Leuco- 
termes) not heretofore reported from the Islands and one new 
genus, Planocryptotermes, The list of species described is as 
follows : 

Genus Kalotermes Hagen sensu restricto. 

1. Kalotermes mcgregori sp. nov. 
G«nus Cryptotermes Banks. 

2. Cryptotermes cynocephalus sp. nov. 
Genus Piano cryptotermes gen. nov. 

3. Piano cryptotermes nocens sp. nov. 
Genus Prorhinotermes Silvestri. 

4. Prorhinotermes luzonensis sp. nov. 

5. Prorhinotermes gracilis sp. nov. 
Genus Leucotermes Silvestri sensu restricto. 

6. Leucotermes philippinensis sp. nov. 

These species, with the exception of P. nocens, belong to genera 
of widespread occurrence, known from Japan and Formosa to 
the north (except Kalotermes) and from the East Indies, Ceylon, 
and India to the south, and it might well have been predicted 
that such termite species would be found in our Philippine fauna. 
Therefore, the fact that the former collections made by Baker and 
by McGregor as well as my earlier collections failed to bring 
them to light might well cause surprise. The reasons for their 
not having been collected, however, are not far to seek. With 
exception of the Planocryptotermes species, they are not among 
our common forms. Nor are they conspicuous, since none of 
them builds mounds or exposed nests ; nor do the first five species 
build exposed galleries, while the last one seems to be a rare 

I wish to take this opportunity to thank Mr. R. C. McGregor, 
ornithologist of the Bureau of Science, for his never-failing in- 


24 The Philippine Journal of Science 1921 

terest, his aid in collecting, and for affording me, while acting 
director, the facilities of the Bureau in making drawings, taking 
photographs, etc. To Dr. Sanji Hozawa, of the Japanese Impe- 
rial Plant Quarantine, and to Dr. Masamitsu Oshima, director of 
the Government Institute of Science of Formosa, both expert 
termitologists, I wish to express my gratitude for splendid sets of 
comparative material, including many cotypes or autotypes of 
Japanese and Formosan forms and, in the case of Doctor Oshima, 
for autotypes of many of his Philippine species. This material 
has been and will continue to be of great value in determining 
our Philippine termites. 

All of the species described in this paper, like the other species 
of the lower families of the order, present a rich protozoan fauna 
within the hind gut. Prof. C. A. Kofoid, of the University of 
California, who has done much work with the protozoa of Amer- 
ican termites and who with his staff is entering upon a com- 
parative study of these specialized forms, has kindly consented 
to work up those found in our termites, and I am sending him 
m^aterial as rapidly as is practicable. The results of these 
studies should throw an interesting light on our classification, 
and a knowledge of the ^'parasites'' may prove of real value in 
classifying the species and properly grouping the genera. 


After careful study and correspondence with various students 
of termite classification I have decided to make those changes in 
generic and family names which, as Banks has recently pointed 
out, will be necessary if we follow strictly the international rules 
of zoological nomenclature. The necessity for having and observ- 
ing such a set of rules is so obvious and has been so thoroughly 
discussed that I need not defend my action in this matter. The 
changes are inevitable, and the sooner we accept and use them 
the less difficulty will there be and the sooner will we arrive at 
a firm basis for our nomenclature. It was only after long 
hesitation and with great regret that I felt myself forced to 
adopt these changes which must for the time result in such 
an unfortunate confusion of generic names of long standing. 
I find that others of the younger workers in the group have 
passed through the same attitude of mind to arrive at the same 

The tendency is apparent, in most recent publications on this 
group, to do away with the awkward tripartite names by raising 

19, 1 Light: Notes on Philippine Termites, II 25 

the subgenera to generic rank wherever possible. Such a change 
seems to me to be conducive to clearness and usableness, par- 
ticularly in the older genera, Kalotermes, Eutermes, and Termes, 
and I shall adopt it in my work on the Philippine termites. 

With the above changes Holmgren's arrangement of families 
and genera is an admirable one, and I shall follow it. It mjay 
be of interest to point out here that in addition to the characters 
already pointed out by Holmgren as separating the three higher 
families it is very significant that the protozoan faunae of the 
guts of the three groups are characteristic. I find none of 
the polyflagellate protozoa (Hypermastigina) in the gut of any 
of the Termitidae (Metatermitidse of Holmgren) , and those of the 
Rhinotermitidse (Mesotermitidse of Holmgren), while very simi- 
lar in all the genera of that family, are quite different from those 
found in the Kalotermitidae (Protermitidse of Holmgren). This 
is significant in connection with the position which Banks ^ gives 
to Leucotermes in his classification. 


While measurements made from a small range of specimens 
must not be considered as fixing the variational range for a 
species and must be used with caution, such measurements are 
of undoubted value in the determination of species in a group 
where specific lines are by no means easy to draw, and I shall 
as a rule accompany my descriptions by a set of such measure- 

There has been a considerable degree of carelessness on the part 
of some workers in furnishing the details necessary for an 
intelligent use of specific descriptions. Measurements are given 
for body length, head length, head length without mandibles, 
pronotum length, etc., without making it clear just what such 
measurements mean. If systematic work is to accomplish any- 
thing worth while the forms of animals should be so described 
that they may be recognized by other investigators — not only the 
specialist in the group, but the biologist interested in the study 
of animals from other points of view, or even the layman 
desirous of knowing the common forms of life about him. But 
far too many systematic descriptions seem to be written for the 
specialist only, and they are often of little value to him in the 
absence of type material. May I go further and speak from 
the experience gained in entering a new systematic field? The 
needless use of terms of limited application should be avoided. 

^ Banks, N., and Synder, T. E., Bull. U. S. Nat. Mus. 108 (1920) 75. 

26 The Philippine Journal of Science 1921 

Such terms as are used should be made clear. References to 
literature should be given whenever available. Descriptions 
should be comparative. Specific diagnoses are often very- 
valuable. Systematic discussions should point out which are 
the nearly related species and in what particulars the new 
species differs from these nearly related forms. All this will 
add little to the labor of the systematist who has such details 
at his immediate command, and will add immensely to the use- 
fulness and value of his work, not only to the general student 
but to the systematist so unfortunate as to lack a wide range 
of comparative material and a complete library. 

To return to our original question : Just what is the meaning 
of many of the measurements used ? For instance, body length ? 
Does it mean from the distal tip of the mandible to the posterior 
tip of the abdomen ? If so, does it mean with the head extended 
forward or in any position in which it may chance to be, and 
with the mandibles crossed or extended? Or does it mean the 
length of the thorax and abdomen ? These are not idle questions. 
They involve a difference of several millimeters in animals less 
than a centimeter in length. In the Macrotermes soldier, for 
example, the head may assume any position, from that in which 
it forms a line with the long axis of the body to one in which 
it forms a right angle with the body, making a difference of 
3 millimeters or more in total length. Again, head length with- 
out mandibles or, in the nasute soldiers, without rostrum is 
a very indefinite measurement unless carefully defined. 

To avoid the difficulties that I have experienced in using 
descriptions I shall define those terms and measurements which 
I expect to use in my future descriptions. Some changes and 
additions will undoubtedly be necessary as the work develops but 
these will be explained as they arise. 


Frontal suture (stem of the Y suture of some authors). — A 
median longitudinal suture dividing the epicranium into two 
equal lateral halves in the region of the vertex. Absent or 
imperfect in most soldiers. 

Transverse suture (arms of Y suture) . — Separates the vertex 
from the frons. Absent or imperfect in most soldiers. 

Clypeofrontal suture. — Separates frons and clypeus. Absent 
or imperfect in most soldiers. 

19, 1 Light: Notes on Philippine Termites, II 27 

Clypeal suture. — Divides clypeus transversely into a distal 
anteclypeus and a proximal postclypeus. Lacking in some 

Labral suture. — Between anteclypeus and labrum. 


The head sclerites are not clearly marked, particularly in 
the soldier, hence the areas or regions referred to are necessarily 
more or less indefinite. 

Labrum ("Oberlippe'' of Holmgren). — Upper lip. 

Lingula (Fuller, 1915) . — Anterior hyaline extension of labrum 
found in certain soldiers (of Macrotermes, for example). 

Anteclypeus ("Clypeoapicale" of Holmgren). — Distal region 
of clypeus betv^een labial and clypeal sutures. 

Postclypeus (*'Clypeobasale" of Holmgren). — Proximal region 
of clypeus, between clypeal and clypeofrontal sutures. 

Frons, front (*'Transversalband'' of Holmgren). — The region 
bounded posteriorly by the transverse suture, anteriorly by the 
clypeofrontal suture, and laterally by the antennal carinae. Not 
at all or imperfectly defined in most soldiers. 

Frontal area (Fuller, 1915). — Forehead ("Stirn"), including 
frons and clypeus where transverse suture and clypeofrontal 
suture are both obsolete, as is the case in most soldiers. Region 
between fontanelle and clypeal or labral suture, since fontanelle 
is typically located at junction of frontal and transverse sutures. 

Vertex, — The top of the head corresponding to epicranial 
region of insects whose head sclerites are well defined. 

Occiput (occipital region). — "An indefinite area forming the 
convex caudal extremity of the head.'' (Fuller, 1915.) 

Gense. — Sides or cheeks of head. An indefinite area not deli- 
mitated in termites. 

Ventral gense. — ^Ventral surfaces of the head lateral to the 
gula including postgense which are not delimitated. 

Gula (*'menton" of Bugnion, "submentum" of Holmgren). — 
A distinct median ventral sclerite, articulating anteriorly with 
the labium. 


Fontanelle. — A foramen in the epicranium, usually in the 
frontal suture at its junction with transverse suture. 

Fontanelle plate, — The region of the frontal gland marked 
externally as a thickened or darkened area. 

28 The Philippine Journal of Science 1921 

Antennal fossx ("Antennenvertiefungen'' of Holmgren). — 
The depressed lateral areas from which the antennae arise. 

Antennal foveolx (Fuller, 1915). — The pits from bottom of 
which the antennae arise. 

Margins of antennal foveolx, — Chitinous margin of antennal 
pits which is usually thickened, often raised, extended, or 

Antennal carinas (^'Antennenleisten'' of Holmgren). — The 
ridges above, that is medial to, antennal fossae. 


Body length. — By this I mean, unless otherwise stated, the 
distance in a straight line from that part of the head, with 
exception of the antennae or palpi, which happens to be most 
distal (with soldiers usually the tips of the mandibles, and with 
workers or adults the clypeus or labrum) to the posterior tip 
of the abdomen. As this measurement varies greatly with the 
position of the head, method of killing, preservation, etc., it 
should be used with caution in differentiating species. 

Body length without head, — From the anterior edge of pro- 
notum in the midline to the posterior tip of the abdomen. In 
using this and other measurements of body length it should be 
kept in mind that specimens preserved in alcohol often undergo 
a very distinct swelling, heavily chitinized regions becoming 
widely separated, as a result of which body length becomes con- 
siderably increased over that normal for the species in life. 

Head length. — In the soldier this is the distance from the pos- 
teriormost part of the head to tip of the mandibles. This dis- 
tance is usually measured with the head removed from the body 
and lying flat, in which case it is from the most posterior visible 
portion of the head in the midline to the tip of the mandibles ; or, 
if these are crossed, to a line from their anteriormost point 
making a right angle with the long axis of the body. It may be 
measured with the head lying on its side from the posterior line 
of the head to the distal tip of the mandibles. In the adult this 
is the distance from the posterior border of the head to the 
most distal part, usually the labrum. Here again we have a 
measurement which varies greatly in some species with the 
change in position of the mandibles, and it should therefore be 
used with caution. 

Head length without mandibles. — Measured, in soldiers, from 
the posterior line of the head to the labral suture, with the head 

19, 1 Light: Notes on Philippine Termites, II 29 

lying flat; if measured with the head on one side, from the 
external articulation of the mandibles to the posterior margin 
of the head. 

Head width. — Measured at the widest point, including eyes 
when present. Considerable confusion has arisen from a care- 
less use of this measurement! 

Fontanelle index. — Distance from the posteriormost part of 
the head in the midline to the fontanelle divided by the length 
of the head without mandibles. I plan to use this value in 
certain species because of the indefiniteness which I have en- 
countered as to the meaning of such statements as: "Fonta- 
nelle at middle of head," "Fontanelle in front of the middle of 
the head," etc. 

Pronotum width. — Measured at the widest point. 

Pronotum length. — Measured in the midline and hence the 
minimum length in species with notched pronotum. I suspect 
that this term is used by some writers, without explanation, to 
mean maximum pronotum length. 


Protermitidae Holmgren. 

Genus KALOTERMES Hagen sensu restricto 

Subgenus Calotermes sensu stricto Holmgren. 


Adult. — Median vein of the forewing runs parallel to the 
cubitus and midway between it and the radial sector, simple or 
branched. Antennae with 16 to 19 segments. 

Soldier. — Head relatively large, elongate, arched, gradually 
flattened anteriorly ; mandibles large, toothed but unsymmetrical, 
all femora enlarged. Antennae with 13 to 18 segments, the 
third typically enlarged, modified, and highly chitinized. Similar 
to soldiers of Neotermes. 

The species of Kalotermes are to be found living in the dead 
branches of living trees, in the dead wood of hollow or injured 
trees or, in some cases, in or very near the live wood. They have, 
therefore, the same habitat as the species of the closely related 
genus Neotermes. They form small colonies of at most a few 
hundred individuals consisting chiefly of larvalike "workers,'' 
a few nymphs of supplementary reproductive forms, and a few 

Kalotermes seems to have its greatest development in the 
Nearctic Region where Banks has reported nine species. It 

30 The Philippine Journal of Science 1921 

seems to be replaced in the main in the Oriental Region by the 
species of Neotermes and Glyptotermes. The species described 
here is the first species of the genus Kalotermes reported from 
the Philippines, and the second from the Oriental Region, the 
only other species being K, indicus (Holmgren) reported from 
Macassar and Siam. 

Kalotermes mcgregori sp. nov. Plate 1, figs. 1 and 2, text fig. 1. 

Types, — Short-headed soldiers (from No. 188 of general col- 
lection), long-headed soldiers, **workers," and nymphs (from 
No. 289 of general collection). No. 24 in type collection. 

Cotypes. — No. 188 in general collection {McGregor and 
Light) , Culi Culi, Rizal Province, Luzon, near Manila, October 3, 
1920 ; No. 289 in general collection (McGregor and Light) , Culi 
Culi, November 19, 1920, same colony as No. 188 ; No. 339 (Mc- 
Gregor and Light) f Rosario, Batangas Province, Luzon, De- 
cember 25, 1920. 


Body of all castes broad and flat ; thorax long ; head and body 
hairy; antennae of soldiers with 15 to 17 segments, the third 
heavily chitinized and twice as long as the second. Dorsal and 
lateral margins of antennal foveolae projecting. Pronotum long 
and very broad, much broader than the head and strongly arched, 
deeply concave anteriorly, its anterolateral regions projecting 
over the head. Abdominal terga of soldier somewhat chitinized. 
Living in the trunk of ipil-ipil (Leiicaena glauca Benth.) . 


Adult, — Unknown. Well-developed wing pads show (March) 
the median to lie parallel to and midway between the radius 
sector and the cubitus. 

Soldier, — Head shading from yellow posteriorly to chestnut 
anteriorly, mandibles black ; antennae brown proximally, shading 
into very light yellow distally. Pronotum, mesonotum, and 
metanotum light brown; abdominal tergites, tibiae and tarsi of 
legs yellow with a faint brownish tinge or light brown, femora 

Head, body, and legs covered with a dense growth of subequal 
microscopic hairs; head short, thick, and directed somewhat 
downward, flat below, rounded liaterally and above, converging 
slightly at both ends, bluntly rounded posteriorly. Frons 
rather precipitate, slightly concave in central region. A few 
soldiers have longer heads with straight sides converging but 

19, 1 Light: Notes on Philippine Termites, II 31 

little at either end and are marked by the apparent absence of 
eyes and by the presence of a hyaline spot near the antennae (see 
text fig. 1). Mandibles (see Plate 1, fig. 1) short, stout, with a 
very distinct upcurve, and slightly incurved tips; their outer 
surfaces show a low basal hump, a slight concavity in the center, 
and a convexity near the distal end ; left mandible a little longer 
than right, with three teeth on cutting edge ; distal tooth double 
and extended distally, second triangular, somewhat truncated 
distally with a low posterior projection confined to the dorsal 
region of the mandible ; basal tooth large and bluntly triangular ; 
right mandible with two triangular teeth, with short distal and 
long proximal faces; proximal region of mandible roughened. 
Labrum about twice as broad as long, reaching to the anterior 
border of lower tooth of right mandible, parallel-sided, with 
slightly rounded anterolateral comers and a slightly convex an- 
terior margin bearing a number of bristlelike hairs. 

Antennae with 15 to 17 segments, first segment large, cylin- 
drical, and nearly hidden from above by the projecting dorsal 
margin of the antennal foveolse; second short and cylindrical, 
third large and heavily chitinized, obconic with a proximal diam- 
eter less than that of the second segment ; next six obconic but 
short and thick ; more distal i^egments thickly clavate and lightly 
chitinized, apical segment oval, white. Eye hyaline and sepa- 
rated by less than its diameter from edge of antennal foveola 
(not discernible in long-headed soldiers) ; gula short and broad, 
anterior region but little less than twice as wide as narrowest 
portion. Legs short, femora swollen ; pronotum large and con- 
siderably broader than head, much arched, making nearly a 
semicircle in transverse section ; anterior border not notched in 
midline but deeply concave, the rounded anterior corners project- 
ing far over the posterolateral regions of head; median long- 
itudinal line distinct; broadest point of pronotum in line with 
the center of anterior border; lateral border receding to meet 
the nearly straight, weakly arcuate, posterior border; anterior 
margin slightly upraised and marked by a dark brown edge; 
mesonotum and metanotum short, mesonotum about two-thirds 
as long as pronotum, with notched posterior border, metanotum 
shorter than mesonotum. Body distinctly flattened, thoracic re- 
gion as long as the abdomen in dorsal view; abdominal tergites 
chitinized. Practically all soldiers collected show wing pads 
varying in size. 


The Philippine Journal of Science 

Fig. 1. Kalotermes m4;gregori sp. nov. Outline drawing of head and pronotum of long- 
headed soldier. Note absence of distinct eyespots and the presence of curious hyaline spots 
in anterolateral region. 

19, 1 

Light: Notes on Philippine Termites, II 
Measurements of Kalotermes mcgregori sp. nov.y soldier. 


Body length 

Body length, without head 

Head length - 

Mandible length, dissected: 


Right - 

Head length, without mandibles _ 

Head width. 

• Pronotum length -. 

Pronotum width _ 













''Larvsey — Large, broad, and thick. Antennse with 11 to 17 
segments; when 17, segments 2 and 3 incompletely separated; 
other segments short, thickly clavate, with thick distal and 
narrow proximal ends, or suborbicular. 


While it is difficult to determine the generic position of the 
species of Kalotermes in the absence of the adult, I feel 
that there can be little doubt in the case of the present species. 
The short legs with swollen femora, the large and heavily chiti- 
nized third antennal segment, the presence of distinct wing 
pads and, finally, the distinct difference in shape of body, degree 
of chitinization, size and shape of pronotum, etc., which dif- 
ferentiate it from the common species of Neotermes, make it 
practically certain that we have in this species a representative 
of the genus Kalotermes, which is here reported from the Islands 
for the first time. Were it not for these striking differential 
characters one might well hesitate to report a Kalotermes species 
in the absence of the winged adult, in view of the absence of any 
species of this genus in the known termite fauna of Formosa 
and Japan to the north and the East Indies, Ceylon, etc., to the 
south, the only oriental species being K. indicics (Holmgren), 
known only from the adult. An exiamination of the venation 
of the wing pads of "workers'' collected recently confirms 

34 The Philippine Journal of Science im 

my diagnosis, as the median runs parallel to and midway be- 
tween the radius sector and cubitus.^ 

The protruding margin of the antennal foveolse of the soldier, 
the very large, characteristically shaped pronotum, the presence 
of wing pads, and the characteristic toothing of the mandibles 
suffice to differentiate this species from other species of the 

I have named this distinct species in honor of Mr. R. C. 
McGregor, ornithologist of the Bureau of Science, who helped 
me to collect it and whose aid and interest have to a great extent 
made possible the rapid collection of local Philippine termites. 


This species was found living in tunnels very close to, if not 
actually within, the live wood of a small leguminous tree. Leu- 
caenxi glauca Benth., known locally as ipil-ipil (Tagalog). In- 
troduced from America, this plant is widespread about towns 
and country dwellings where its rapid growth, which enables 
it to drive out cogon grass, and its usefulness for firewood 
and fence posts make its propagation worth while. The very 
interesting question arises at once as to whether this species is 
found in other and native trees or is confined to this plant and, 
if so, whether it was introduced with the plant and is, therefore, 
an American species, or whether it has become adapted to this 
habitat since the introduction of the plant here. It seems very 
unlikely that plants large enough to harbor these termites were 
brought here, but it is by no means beyond the range of pos- 
sibility. A review of the American Kalotermes species shows 
this species to be most nearly related to K. jouteli Banks, which 
it resembles in the shape of the head, the toothing of the man- 
dible, the size of the third antennal joint, etc. It differs from 
it in many points, however, such as size and shape of the pro- 
notum, shape of the third antennal joint, the projecting margins 
of the antennal foveolse, etc. It appears to be a new species, 
therefore, w^hether introduced from America or not. 

' Since writing the above I have taken two winged adult Kalotermes 
specimens in Cebu Island. Whether these belong to this or some related 
species cannot be determined until the series is completed in either locality. 
There remains the possibility that the form described here represents the 
soldiers and workers of K. indicus Holmgren known from the adult only, 
but the widely separated habitat makes this extremely improbable. 

19,1 Light: Notes on Philippine Termites, II 35 

It was first found by Mr. McGregor and myself (No. 188) 
on October 3, 1920, at Culi Culi, Rizal Province, near Manila, 
when two soldiers and numerous "workers'' were collected. 
Later, November 19, more extensive collections were made from 
the same tree, numerous soldiers and very many "larvae" and 
"nymphs" being collected (No. 289). The termites were found 
living in channels deep in the heartwood and were apparently 
rapidly destroying the tree. At the first collection one side of the 
tree, which was dead, contained in numerous tunnels near the sur- 
face large numbers of workers and soldiers of a species of Nasuti- 
termes, whose tunnels were separated internally by very thin 
walls from those of the Kalotermes colony. Such associa- 
tions, whether chance relations or not, are very common. In 
the case of Prorhinotermes luzonensis, described below, three 
species were involved, Prorhinotermes, a Hospitalitermes species, 
and a Neotermes species. Prorhinotermes gracilis was also 
found in close association with a Neotermes species. 

At the second collection many "nymphs" were found, some 
large with large wing pads, others small with but slight begin- 
nings of wing pads but easily distinguished by their opaque white 
color as contrasted to the dirty white color of the posterior ab- 
dominal region of the "workers", many of which were as large as 
the largest nymphs but showed no wing pads. It was only at 
this collection that the long-headed soldiers were taken. 
Were it not for their presence in both colonies one might sus- 
pect that they represented a different species, so distinctly dif- 
ferent are they from the more numerous short-headed soldiers. 

A second colony, living like the first in ipil-ipil, was found 
by Mr. McGregor and myself on December 25, 1920, while on a 
collecting trip to Batangas, in the municipality of Rosario, Ba- 
tangas Province, some 70 kilometers from the first colony. The 
finding of this second colony in the same tree species makes it 
seem that the species is a regular inhabitant of this tree, whether 
it is able to live in others or not. One or more Neotermes 
species also inhabit the ipil-ipil as they do also the guava, 
the cacauate (Gliricidia maculata HBK) and the ciruelas 
(Spondias purpurea Linn.) ; but many examinations of the last 
three trees, while producing large collections of Neotermes, have 
failed to show Kalotermes mcgregori or any related species. 

On March 31,i 1921, Mr. McGregOT and I again visited the 
colony at Culi Culi and cut down the tree for further study in 
the laboratory. The termites had been driven by the dry wea- 
ther to the deeper and damper portions of the tree, particularly 

36 The Philippine Journal of Science 1921 

to the heartwood, where they were running longitudinal tunnels. 
The forms which I had formerly distinguished as ^'workers'' as 
contrasted with the nymphs all bore large wing pads, which 
would seem to bear out the belief that there is no definitely dif- 
ferentiated worker caste, and that the opaque so-called nymphs 
are probably early stages of supplementary reproductive forms. 

I was greatly pleased on examining the venation as seen in 
these wing pads to find that the median runs parallel to and 
midway between the radius sector and the cubitus, thus con- 
firming my diagnosis of this as a species of Kalotermes. 

Material for a study of the protozoan fauna of this species 
has been sent to Professor Kofoid. 

Subgenus Cryptotermes (Banks) Holmgren. 


Adult. — Median vein bends up to unite with the radial sector 
beyond the middle of the wing. Wing iridescent. Antennse of 
14 to 16 segments. 

Soldier, — Head short, high, and thick, bilobed anteriorly, with 
a vertical frontal area containing a distinct cavity. Mandibles 
short, humped basally, bent near the middle, weakly toothed or 
untoothed. Antennse of 9 to 13 segments, the third not es- 
pecially long. Pronotum with strongly concave anterior mar- 
gin, not toothed. Styles reduced. 

The genus Cryptotermes comprises species from all parts of 
the world. They are typically house termites, living in boards, 
furniture, etc., in houses or, more rarely, in dead wood of trees. 
They are extremely ubiquitous; there is hardly a house in the 
Islands but harbors these little wood destroyers or those of the 
very closely related genus Planocryptotermes. Haviland points 
out a similar condition in Borneo, and it is probably true in the 
entire tropical Indian and Malayan regions. Their presence is 
usually manifested by the piles of little impressed faecal pellets 
which they drop from apertures in the board they are attacking. 
There are apparently a number of Philippine species with this 
habit. Their collection is made difficult by the necessity of 
removing or destroying boards, furniture, etc., to get at the 

Cryptotermes cynocephalus sp. nov. Plate 2, figs. 1 and 2. 

Types, — Adults, soldier, larva, and nymph. No. 25 in type col- 
lection. Soldier, larva, and nymph from No. 67 of the general 

19, 1 Light: Notes on Philippine Termites, II 37 

collection, dealated adult, No. 433 of general collection, winged 
adult from No. 443 of general collection. 

Cotypes. — No. 67 (del Rosario), Manila; No. 433 (Light), 
Manila; No. 443 (Gamboa), Manila; No. 448 (Aguila), Manila, 
in general collection. 


Adult. — Very small, less than 5 millimeters long without 
wings; slender, dark, with narrow, dark wings; antennse long, 
with 13 to 15 segments, third smallest. 

Soldier. — Very short, about 3.25 millimeters long; head short, 
thick and high, bulldoglike, frontal region very strikingly de- 
veloped with distinctly bilobed dorsal margin, anterior cavity 
deep ; a distinct median dorsal cavity present ; anterior and ante- 
rodorsal margins of the antennal foveola extended anteriorly to 
form a flat projection (spine) with rounded tip. 

Larvae and nymphs. — Small and slender. 


Adult. — Head flat, longer than broad; disregarding the eyes, 
nearly parallel-sided, sparsely haired; eyes not large or promi- 
nent, ocellus elongate in an anterodorsal direction, in contact with 
the eye in front of its middle. Antennae more than twice as 
long as width of head, 13 to 15 segments, first segment large and 
cylindrical, second smaller and cylindrical, third smallest, thickly 
obconic, fourth to sixth gradually increasing in size, thicker 
than third, and with rounded distal ends, eighth to thirteenth 
(when 14) increasingly long and with thicker distal ends, last 
as long as thirteenth but narrow and oval in shape; antennae 
with scattered larger hairs and a dense coat of short straight 
hairs. Labrum swollen, yellow; anteclypeus white; postclypeus 
brown, about as long as anteclypeus, remainder of head brown; 
pronotum brown, arched, narrower than the head (with eyes), 
concave anteriorly, sides somewhat rounded, posterolateral 
corners bluntly rounded, posterior margin nearly straight, 
slightly notched at center. Anterior wing scales brown, much 
longer than the posterior pair and reaching to or beyond the 
middle of the latter; abdominal tergites dark brown, metano- 
tum yellow, giving the appearance of a transverse light band 
in dorsal view, abdominal sterna brown with a narrow median 
light area. 

Wings slender, cloudy gray, iridescent; median and cubitus 
very slightly chitinized except at base in some specimens, costs 
and radius sector heavily chitinized, gray-brown; all veins and 

38 The Philippine Journal of Science 1921 

branches as well as areas between the veins marked by papilla- 
like projections; base of anterior wing twice the width of base 
of posterior wing ; f orewing with radius sector and median sep- 
arate at base; radius sector sends five distinct branches, and 
one or more small branches to the costa; median joins radius 
sector beyond middle of wing usually near to origin of third 
branch of radius sector; cubitus with eleven to thirteen 
branches; hind wing with six or seven branches uniting radius 
sector and costa, inner three large; median and radius sector 
usually united for a short distance at their bases ; median joins 
radius sector distally between points of origin of second and 
third branches of latter (in one specimen median runs to end of 
wing!) ; cubitus with about eleven branches which tend to be 
more subdivided than those of f orewing; cubitus of both wings 
bends slightly toward radius sector near level of junction of 
median and radius sector; a few indefinite cross veins unite 
cubitus and radius sector beyond junction of latter with median. 

Measurements of Cryptotermes cynocephalus sp. nov., winged adult. 


Wing length 6.25 

Body, without wings 4.75 

Head length 1.00 

Head width 0.80 

Antenna length 1.65 

Pronotum width 0.75 

Pronotum length 0.45 

Soldier, — Mandibles and anterior region of the head black 
shading into a dull purplish on posterior portion of the head 
and lateral cervical sclerites; antennse, palpi, and distal por- 
tions of the legs light yellow, body segments and proximal 
portions of legs pale purplish brown. Head directed ventrally 
and nearly at right angles to body, short, thick and high, 
broadest anteriorly, suggesting the head of a bulldog (hence the 
specific name) ; anterior surface deeply excavated and extensively 
roughened and sculptured; lateral and dorsal margins of this 
frontal area produced to form a very marked, outwardly and 
forwardly directed flange, deeply notched in the midline, giv- 
ing the head the bilobed appearance characteristic of the 
genus; below, this flange runs laterally on either side to form 
the anteriorly extended posterodorsal margin of the antennal 
foveola, being separated by a groove from the anterodorsal and 
anterior margins of the foveola which form an anteriorly pro- 
jecting, laterally flattened, scalelike spine, whose upper portion 

19,1 Light: Notes on Philippine Termites, II 39 

is overlapped above by the lower portion of the frontal flange; 
below this the ventral gena is extended to form a much smaller 
spine, lying over the mandibular hump. 

Dorsal profile in side view high, domed posteriorly, sunken 
in the middle and elevated anteriorly. Seen from above there 
is a very distinct concavity in front of the middle of the head 
and behind the flange. This is narrow and elongated and, with 
the middorsal notch in the flange, gives the head a distinctly 
bilobed appearance in dorsal view. The surface of the flange and 
the region of the head posterior to it, particularly the sides of 
the dorsal cavity, are distinctly rugose. 

Mandibles very short and strongly incurved, when closed 
protruding for a distance but little more than one-third of the 
anterior surface of the head ; antennae of from 9 to 12 segments, 
longer than the head depth; third segment smaller than the 
second and more heavily chitinized; pronotum strongly arched, 
deeply bilobed anteriorly, rounded laterally and posteriorly, the 
anterior and posterior regions being elevated. 

Measurements of CryptotevTnes cynocephalus sp. nov.y soldier. 


Body length 


Body length, 

without head 


Head length 



to middorsal 





Head width 


Pronotum length 


Pronotum width 


''Workers*' and nymphs, — Small and slender. Considerably 
smaller than those of Planocryptotermes nocens g. et sp. nov. 
which they otherwise resemble. 


This species is characterized by the small size of all castes 
(less even than that of C cavifrons Banks) , the strikingly de- 
veloped and roughened margin of the frontal region of the sol- 
dier, and the very small mandibles. The head profile of the 
soldier resembles that of C. cavifrons, but it differs from the 
latter and resembles C. brevis (Walker) in the roughened con- 
dition of the anterior and dorsal regions of the head. From the 
Japanese form, C kotoensis Oshima, the soldier differs, among 
other characters, in its small, weak mandibles. From C. domes- 
ticus (Haviland), its nearest neighbor geographically, it differs 
in its smaller body size, its smaller and narrower head, its 
smaller mandibles, in that the anterior surface of the head 

40 The Philippine Journal of Science 1921 

makes slightly more than a right angle with the mandibles, 
and in the shape and position of the spine near the antennal 
foveola, which in C. cynocephalus is simply an extension of the 
anterior and anterodorsal margin of the foveola. Specimens 
just received of the Hawaiian CryptotermesAike species, kindly 
sent me by Mr. David T. Fullaway, entomologist of the Bureau 
of Agriculture, show a very striking difference among other 
points in their greater size. From what Mr. Fullaway has told 
me of the venation of the adult I am led to believe that this as 
yet undescribed species must be placed in another genus rather 
than in Cryptotermes. 


House termites, living in the dry, seasoned wood of planks 
and boards of houses, in furniture, in picture frames, etc. The 
specimens on which this species is based were found together 
with those of Piano cryptotermes nocens sp. nov. (see below) by 
Prof. Jose I. del Eosario, of the department of chemistry. Col- 
lege of Liberal Arts, University of the Philippines; they were 
living in boards of his house and were kindly collected for me. 
Whether the two species were in the same boards or not I do 
not know since the material was collected under the impres- 
sion that but one species was involved. This was true also 
of the second colony of this species collected which was found 
living in the same board with P. nocens sp. nov. This colony 
was collected by Cipriano Gamboa on March 7, 1921, and in- 
cluded a few winged adults (No. 443). A third, larger, colony 
(No. 448) was found by Paulino Aguila in the boards of a 
house on March 8, 1921, and contained numerous winged adults. 
This would seem to be very near the time for swarming, as 
many of the adults were fully pigmented and able to fly freely.^ 
Among the hundreds of "workers" and adults, only five soldiers 
were found. These, like those of the colony mentioned above, 
were dead, as the boards had been exposed to the sun. Since 

' Since writing the above, winged specimens of C. cynocephalus have 
been taken at various times during the months of June and July in my 
house in Paco, Manila, which ie badly infested with house termites. These 
adults are never numerous and, strange to say, emerge in the early 
morning rather than at night, as is the habit of Planocryptotermea and 
most other termites. During the latter part of Juiie and the early part of 
July a few of them were to be taken on my window curtain each morning 
and a dealated pair of these, our tiniest adult termites, were commonly 
to be seen in the early morning coursing excitedly over my washstand. 

19,1 Light: Notes on Philippine Termites, II 41 

no extensive collections of house termites have been made here 
or elsewhere, I do not know whether C. cynocephalus and P, 
nocens have the same distribution; nor do we know their re- 
lative frequency. We do know that practically every house in 
the Islands is infested with this or other species of house 

These termites attack isolated boards and are therefore not 
reached by methods which prevent the activity of the much more 
seriously harmful Coptotermes, Leucotermes, and Eutermes 
(Microcerotermes) species, which require a connection with the 
ground or a considerable moi^ure supply. Since the "house ter- 
mites'' reach their future habitat in a winged state, there is no 
way of preventing their presence in tropical regions where it is 
impossible to keep the house closed against their ingress. The 
only methods of combating them would be, therefore, the use of 
treated lumber, together with the prompt removal of any infested 
boards, which would presumably be prohibitive in cost. Their 
presence is usually demonstrated at once by the little piles of 
impressed pellets of faecal matter which they throw out from their 
galleries : during the night, as a rule, but sometimes during the 
day. These little piles of yellow or orange-colored pellets (color 
depending upon that of the wood) are very characteristic sights 
in our houses in the Tropics. These termites are taken by many 
to be beetles because of the larvalike appearance of the worker 
and the curious color and shape of the soldier. Indeed they are 
locally known as gorgojo (beetle) or more commonly bucbuc 
(borer) and with the species of the genus Planocryptotermes 
are the only termites not recognized as such and given the name 
anay by Filipinos generally. 



Imago. — As in Cryptotermes but with as many as 18 ^ segments 
in antennse. 

^ Collections which I made in Cebu and Negros during April and May, 
and further collections in Manila, have produced no new species of house- 
living Cryptotermes, They have shown Planocryptotermes to be common 
and apparently much more prevalent than Cryptotermes and have pro- 
duced another species of Planocryptotermes from Manila, one from Cebu, 
and apparently two from Negros, all closely related to P. nocens. We 
would seem justified in the belief, therefore, that the "house termites," 
whose piles of faecal pellets are to be found in nearly every dwelling in 
the Philippines, belong in great part to the new genus Planocryptotermes. 

' In the adult of an undescribed species from Manila. 

42 The Philippine Journal of Science 1921 

Soldier. — Similar to Lobitermes Holmgren. Head broad, flat 
and smooth, somewhat longer than broad ; forehead nearly ver- 
tical, with notch in middorsal region of border. Antennse with 
11 to 14 segments, third segment not much larger than second. 
Mandibles distinctly toothed. 

Genotype, Planocryptotermes nocens sp. nov. 

I have founded this new genus for the species described 
below (P. nocens) and several others from other parts of the 
Islands to be described later. Calotermes pinxingae (Haviland), 
which Holmgren places provisionally in his subgenus Lobitermes, 
probably belongs with these species. All the data I can gather 
from descriptions and illustrations point to that conclusion and 
a study of the winged form will probably show this to be true. 

I had been led to the belief that the adult of P. nocens would 
agree with that of Cryptotermes for the reasons that Cryptoter- 
mes'like adults had been taken from time to time in numerous 
places known to be infested with Planocryptotermes, and that no 
other Kalotermitinse adult had been taken in these vicinities 
except the tiny form which I have since determined to be the adult 
of C cynocephalus. With so many houses infested with these 
termites it seemed extremely unlikely that the adults had escaped 
notice so long and as I have pointed out above the only adults 
captured showed the Cryptotermes type of wing venation. 

Very recently adults found in colonies of P. nocens and other 
species of the genus have confirmed this assumption, making it 
necessary to establish the new genus for the group, since the 
adults of Lobitermes do not have the Cr'yptotermes venation. 

The soldiers of the new genus are characterizfed by a larger 
size than those of Cryptotermes, by a larger head which is flat- 
tened dorsally and is considerably broader than high and some- 
what longer than broad, by a' more or less pronounced notch 
in the projecting rim of the frontal area, and by the absence 
of any considerable elaboration or rugosity in the frontal area 
which characteristically makes more than a right angle with 
the mandibles, which in turn are longer and slenderer than in 
Cryptotermes and distinctly toothed. They are characteristically 
house termites but are sometimes found in dead limbs of trees. 
Like Cryptotermes their presence is denoted by the piles of little 
faecal pellets dropped from openings in the boards they inhabit 
(see Plate 6). 

19, 1 Light: Notes on Philippine Termites, II 43 

Planocryptotermes nocens sp. nov. Plate 2, figs. 3 and 4 ; Plates 5 
and 6. 

Types, — Adult, soldier, larva, and nymph. No. 26 in type col- 
lection from No. 202 of general collection. 

Cotypes, — No. 39 (del Rosario), Manila; No. 202 (Light), 
Manila; and No. 442 (Gamboa) , Manila, in the general collection. 


Adult. — Antennae long, with 16 segments, rather sparsely 
haired with stiff hairs of two sizes, segments 11 to 15 thickly 
clavate with very slender proximal ends, terminal segment shorter 
and much narrower, ovate; body with wings from 8.5 to 9 milli- 
meters long, without wings from 5 to 5.5 millimeters long; 
pronotum slightly narrower than head. 

Soldier, — Head about 1.25 millimeters long from posteridr bor- 
der to middorsal margin of frontal area, about 1.50 millimeters 
from posterior border to labral suture; about 1.25 millimeters 
wide; head making more than a right angle with mandibles; 
margin of frontal area not strongly developed, bilobed, with one 
deep median dorsal and two slighter lateral dorsal notches ; ante- 
rior concavity shallow; two short "antennal spines," one an 
extension of the anteroventral margin of the antennal foveola, 
the other of the anterodorsal margin. Antennse of 13 or 14 seg- 
ments. House termites. 


Adult. — General color brown above, ventral surface of thorax 
yellow, of abdomen light brown; wings faintly iridescent, light 
transparent brown, anterior border darker opaque brown ; head 
rounded posteriorly, nearly semicircular, longer than broad, with 
scattered spikelike hairs; Y suture visible in center of head; 
labrum small, somewhat swollen, yellow; antennse typically 
with 16 segments, all segments rather sparsely haired with 
scattered, stiff, larger hairs and more numerous smaller hairs; 
first segment cylindrical, second shorter, slightly swollen dis- 
tally, proximally heavily chitinized ; third thickly obconic, about 
as long as second ; fourth to eighth as long as broad, increasing 
slightly in size and becoming more smoothly rounded; ninth 
larger, broadly oval; tenth to fifteenth increasingly long, 
clavate with very slender bases; sixteenth long-oval, shorter, 
and slenderer. 

44 The Philippine Journal of Science 1921 

Measurements of Piano cry ptotermes nocens sp, nov., winged adult. 


Body length, with wings 8.5-9.00 

Body length, without wings 5.0-5.50 

Head length'^ 1.16 

Head width 1.08 

Pronotum width 1.01 

Pronotum length 0.63 

a Posterior margin to middorsal margin of frontal area. 

Soldier. — Mandibles and frontal area of head, with its margin, 
black, the rest of the head shading from smoky brown in front 
to dirty yellow at posterior surface; thorax and legs smoky, 
abdomen light smoky yellow, having a faint purplish tinge. An- 
tennae and mouth appendages light yellow to white. Head mak- 
ing an angle of about 45° with the body, short, broad, and square, 
being only slightly longer than broad and about half as high as 
long and showing scattered hairs, which are smaller and more 
numerous toward the anterior ; anterior surface making an angle 
of more than 90° with the mandibles; posterior border nearly 
straight in dorsal view, with shortly rounded corners ; sides and 
posterior surface of head sloping inward and upward to dorsal 
surface from a line of maximum convexity considerably below 
the mid-horizontal plane of the head; sides of head nearly 
straight and parallel. 

Eyes small, hyaline and circular, lying in a considerably larger, 
circular elevation of the lateral surface of the head, about 
one and a half times their diameter from posterior margin of the 
antennal foveolse, making a distinct lateral projection in dorsal 
view in front of which the sides converge slightly only to diverge 
in the low but distinct margin of the frontal area, so that the 
width of the head from edge to edge of the frontal margin would 
about equal that through the eyes; in dorsal profile the head 
in side view presents a sharp rise posteriorly from the border 
to a point one-third the distance between the posterior border 
and the margin of the frontal area; from this point there 
is a very gentle down curve to near the frontal margin and a 
short but distinct rise to the anterior edge of the projecting 
margin ; dorsal surface with a very small concavity just posterior 
to the middorsal region of the margin of the frontal area ; frontal 
margin less developed in middorsal region where it shows a 
distinct median notch, flanked on either side by a much smaller 
lateral notch and a second very slight notch ; the margin, slightly 
roughened laterally to these notches, curves anteriorly, laterally, 
and ventrally to end in the dorsal margin of the antennal f oveola 

19,1 Light: Notes on Philippine Termites, II 45 

where it is laterally deflected and continuous with the posterior 
margin of the f oveola. On either side of the mouth parts on the 
ventral side of the head is a dark narrow ridge (the edge of the 
ventral gena) running forward and laterally to culminate in a 
laterally flattened spinelike projection ("antennal spine") of 
the anteroventral margin of the antennal f oveola ; above this the 
anterior margin is reduced, giving the appearance of a deep notch 
internal to which the anter odor sal margin projects as a broader, 
still more prominent "spine/' 

Lying between and above the antenna and the mandible of each 
side is a pair of rounded elevations, the outer slightly more dorsal 
than the inner. Concavity of frontal area shallow and its sur- 
face smoothly rugose ; mandibles distinctly thickened basally with 
a lateral hump, distal three-fourths slender in side view and con- 
siderably flattened dorsoventrally ; left mandible less curved than 
right and bearing three teeth, the distal two small and the prox- 
imal one long and low; right mandible more strongly curved, 
with two teeth, the distal one large, with distal edge at right 
angles to mandible and proximal edge long, making a very obtuse 
angle with surface of mandible ; proximal tooth of right mandible 
low and inconspicuous ; large distal tooth of the right side flts in 
between the two distal teeth of left mandible when mandibles are 
closed, tip of the right mandible crossing under that of the left 
which projects beyond it; labrum white in color, projecting over 
the proximal half of the opened mandibles, narrow, converging 
distally with a distinct rounded point bearing two long, upcurved 
hairs at its tip with several smaller hairs just posterior to them. 
Antennae considerably longer than height of head, 13- or 14« 
segmented ; when 13, the third obconic and nearly as long as the 
second; when 14, the third segment divided to form a short 
obconic third and a short disk-shaped fourth; flrst segment cylin- 
drical, longest and thickest; first, second, and third heavily chiti- 
nized ; first one or two beyond the third less so and the remainder 
very lightly so ; first one or two segments beyond the third often 
nearly disk-shaped, the three beyond these become increasingly 
long, the proximal end more distinctly narrowed and stalked, the 
distal end more rounded (that is, spherical) beyond which, with 
exception of last, they are similar and may be described as short, 
thick clubs with narrow stalks and thick rounded distal regions ; 
last two segments oval, somewhat longer and considerably nar- 
rower than preceding segments, the last only about two-thirds as 
wide as the next to the last. Gula small, weakly chitinized and 

46 The Philippine Journal of Science 1921 

about half as wide as long. Pronotum narrower than head, 
arched, about half as long as broad, and distinctly elevated ante- 
riorly, anterior border deeply concave, thickened, edged with 
black and slightly rugose, anterolateral corners rounded laterally, 
marked anteriorly by a distinct notch and by rugosities ; lateral 
margins nearly straight, converging toward the posterolateral 
comers which round into the slightly convex posterior border. 

Measurements of Planocryptotermes nocens sp. nov., soldier. 


Body length "^ 4.25 -4.60 

Body length, without the head 3.50 -4.10 
Head length: 

Posterior margin to middorsal margin of frontal 

area 1.25 -1.35 

Posterior margin to labral suture 1.50 

With the mandibles 2.20 -2.25 

Head width 1.12 -1.25 

Pronotum width 1.00 -1.00 

Pronotum length 0.575-0.625 

» Made from preserved specimens. Possibly much longer in life. 

''Larva/' — Reaching a length of about 5 millimeters. One of 5 
millimeters shows 12 segments in the antennse. Slender, thorax 
much slenderer than abdomen which is long, swollen posteriorly, 
and colored a yellowish brown by the wood particles and protozoa 
of the gut. A larva of 1.85 millimeters shows antennse of 10 
segments, segments 3, 4, 5, and 6 being rudimentary. 

Nymph. — Similar to "larva" in general appearance but reach- 
ing larger size; having wing pads in various stages of develop- 
ment and distinct, gray, compound eyes. One about 6 millimeters 
in length shows 17 segments in the antennse, 4 of them still 


This species differs from Calotermes pinangae (Haviland), the 
only other described species of the genus, among other points, 
in the toothing of the mandible and in the greater length and 
breadth of the head and pronotum. 


House termites, living in boards in houses, furniture, picture 
frames, etc. More extensive collecting is necessary to deter- 
mine the relative prevalence of the different species of this 
genus. My collection contains several closely related new species 
which I plan to describe in a future number of these notes. 

19.1 Light: Notes on Philippine Termites, II 47 

So far this species has been collected only in Manila and only 
a few times here. Adults of this species were collected by me 
(No. 15) in June, 1920, with numerous dried insects in a hang- 
ing lamp shade at my former house in Ermita, Manila. I had 
noted the appearance of this form on June 12, which points to a 
protracted or irregular swarming on the part of this species. 
The second collection was, as noted above, by Prof. Jose I. del 
Rosario from boards in his house, with C. cynocephalus ; the third 
by me from boards and moulding of a case for birds' eggs hanging 
on a cement wall in the laboratory. The fact that no other 
species of the genus has been found in boards in Manila leads to 
the belief that it is the common species, at least in this locality. ' 
P. nocens, like C. cynocephalus, forms small colonies, eating the 
dry wood, without any direct connection with the ground or any 
external source of moisture. Most of the specimens described 
were found living in the wood of a box containing an exhibit of 
birds' eggs. The box had been hanging on a cement wall sur- 
face in the laboratory for several years without being in contact 
with any other wood. The assistant who hung the box tells 
me that there were signs of termite work when it was hung. 
This would seem to imply that the colony found had been in 
the wood for some years. Since there were less than a hun- 
dred specimens collected, of which but two were soldiers, 
we can get some idea of how slowly such colonies develop. 
Since, also, the wood was by no means all destroyed we can 
get an idea of how slowly they work (see photographs of work. 
Plates 5 and 6). Since writing the above an examination of 
a part of this colony (March, 1921) shows that many of the 
'^workers'' as well as the white nymphs are developing wing 
pads. One soldier and numerous workers were found in boards 
of the house of Prof. Jose I. del Rosario, where in association 
with Cryptotermes cynocephalits they were attacking only the 
boards of white lauan (Anisoptera thurifera Blume), a com- 
paratively soft wood, and avoiding the harder molave and ipil, 
which are nearly termite proof. 

^ Since writing the above a very distinct species has been taken in 
Manila, one of the five species yet to be described which were mentioned 
above. However, winged adults of P, nocens have been found commonly 
about the lights during June and July, and adults of the other species have 
not, and there seems no reason therefore to change our belief that P. nocens 
is by far the commonest house termite of this locality as other closely 
related species of the same genus seem to be in Cebu and Negros. 

48 The Philippine Journal of Science 1921 


Mesotermitidse Holmgren. 

Arrhinotermes Wasmann. 


Adult. — Head broadly egg-shaped, nearly circular. Clypeus 
much broader than long, swollen. Antennae with 19 to 22 seg- 
ments. Pronotum narrower than the head. Wing membrane 
weakly haired, strongly reticulate. The median of both wing 
pairs arises from the cubitus or is lacking or arises from the 
radial sector in hind wing (P. luzonensis!) . 

Soldier. — Head distinctly narrowed distally. Compound eyes 
present, distinct or vestigial. Fontanelle distinct. From the 
fontanelle there runs forward a more or less distinct channel. 
Fontanelle gland large, extending far backward into the body. 
Antennae of 16 or 17 segments. 

Worker. — Clypeus rather large. With or without distinct com- 
pound eyes. 


In 1902 Wasmann described the genus Arrhinotermes for a 
new species, A. heimi from Ceylon, based on adults only. In 
an appendix to the same article he describes A. oceanicus, based 
on adults from Cocos Island. Holmgren (1911) points out that 
A. heimi Wasmann is apparently a Coptotermes, and Bugnion 
in 1910 had reported the adults described by Wasmann as A. 
heimi to be nothing else than the adults of Coptotermes tr avians 
(Haviland). As Holmgren points out, therefore, according to 
the rules of nomenclature, A. heimi being considered as the type, 
Arrhinotermes becomes a synonym. Banks (1920) replaces it 
by Prorhinotermes Silvestri (1909). Holmgren, however, in 
view of the fact that A. oceanicus Wasmann described in the 
appendix to the same article is a true Arrhinotermes, retains 
that generic name with A. oceanicus Wasmann as the type. 
If we are to follow the rules of nomenclature, Arrhinotermes 
must be considered a synonym and I am, therefore, following 
Banks in this matter. 

The genus Prorhinotermes seems to be peculiarly an island 
genus and while widely distributed is represented by few 
species and those apparently closely related. So far as I am able 
to ascertain, no other region has yet produced two species of this 

^^' 1 Light: Notes on Philippine Termites, II 49 

genus. The American species is P, simplex (Hagen), found in 
the East Indies and Florida ; the Formosan species is P. japonicus 
Holmgren; Ceylon has P, flavus (Bugnion and Popoff) ; Samoa, 
P. inopinatiis Silvestri; Krakatoa, P. krakataui Holmgren; 
P. oceanicus Holmgren is found in Cocos Island; and P. was- 
manni Holmgren, in Costa Rica. In view of this peculiar distri- 
bution of the genus it was to be expected that the Philippines 
would show at least one Luzon species and perhaps others from 
different islands. The facts in the case are a good example of 
the surprises which await the termite collector in the Philippines. 
This genus was not encountered until collections were made on 
the Manila-North Road where, near the Bulacan-Rizal boundary, 
three soldiers of P. luzonensis sp. nov. were found, without 
workers, in a stump. Later collections showed, within 50 meters 
of this, a small complete colony of P. gracilis sp. nov. (see below) , 
with a dealated adult living in the hollow end of a branch of a 
guava tree. Still later collections produced from a similar 
situation, within a kilometer or two of the spot, the large colony 
with many winged adults on which the new species P. luzonensis 
is based living in close relation with a Hospitalitermes species 
and a Neotermes species. Extensive collections since that time 
have shown no other colonies in Luzon, although a few winged 
adults have been taken. ^ 

Frorhinotermes luzonensis sp. nov. Plate 3, figs. 1 and 2, text fig. 2. 

Types. — Winged adults, large and small soldiers, workers, and 
nymphs, No. 27 in the type collection (from No. 205 in the gen- 
eral collection). 

Cotypes. — No. 97 (McGregor and Light) , Rizal-Bulacan boun- 
dary, September 13, 1920 ; No. 205 and mixed with No. 209 in 
the general collection (McGregor and Light), Rizal Province, 
about one kilometer from the Rizal-Bulacan boundary on the 
Manila-North Road, October 4, 1920; No. 436, near No. 205 
(McGregor and Light), March 27, 1921. 


Adult. — Length, 7 to 8 millimeters, with wings, 11 to 12 milli- 
meters; anterior wing, 9 millimeters long. Antennae with 18 to 
21 segments. Pronotum 1.50 millimeters broad by 0.75 to 
0.85 millimeter long. Wings in appearance like those of P. flavus 

' Collections recently made in the central islands of Cebu and Negros 
show one or more species of Prorhinotermes to be common in the former 
island and fairly so in the latter. 

180365 4 

50 The Philippine Journal of Science 1921 

(Bugnion and Popoff ) and in venation like those of P. inopinatus 
Silvestri but extremely variable. Median and cubitus more 
or less completely united in anterior wing, tending to separate 
and reunite, forming a closed cell; median of hind wing often 
arising from radial sector or from branches from both radial 
sector and cubitus. Radial sector thickened, particularly in 
outer third of each wing, and united with the costa by 8 or 9 
short, thick, cross veins. All veins marked externally by tiny 
papillae. Gula much broader than long, rounded posteriorly. 
Toothing of mandible like that of P. japonicus and P, inopinatus, 
not like that of P. flavu^. 

Soldier, — Large soldier with long head converging distinctly 
anteriorly, somewhat like P. flavus but with much longer body 
and longer maxillary palpi. Antennae of 18 or 19 segments. 
Head with mandibles, 2.75 to 3 millimeters long; head width, 
1.50 to 1.65 millimeters. Small soldier with head 2.50 milli- 
meters long with mandibles, and about 1.40 millimeters wide. 


Adult. — Head, body, and legs yellow shading into brown in 
older individuals; head, base and anterior border of wings 
and posterior abdominal tergites often smoky. Wings diapha- 
nous with the exception of costa and radial sector which are a 
grayish yellow becoming smoky near the base. Body more or 
less rounded, which, together with its light color and diaphanous 
wings, gives this termite an appearance quite different from that 
of most adult termites; 8 millimeters long, with wings 11 to 
12 millimeters long, anterior wing 9 millimeters long. Head 
broadly egg-shaped, somewhat flattened behind and narrowed 
in front, slightly longer than broad (1.55 by 1.40 millimeters) ; 
posterior border rounded, surface flattened, slightly concave. 
Fontanelle small but distinct, frons incompletely delimitated, 
rising to meet the much-swollen postclypeus which is distinctly 
divided in the midline; postclypeus more than twice as broad 
as long and showing in front of fontanelle a shallov/ channel 
outlined in dark brown; anteclypeus small, white, four times 
as broad as long; labrum large, swollen, with four apical hairs 
in two lateral pairs ; labrum a little longer than clypeus. Ocelli 
very near the compound eyes, small, indistinct, hyaline, an 
elongated oval, long axis nearly parallel to long axis of head. 
Antennae with 18 to 21 joints, first large, cylindrical, heavily 
chitinized, second shorter, narrow and cylindrical, third and 
fourth smallest, disk-shaped, the others orbicular to broad-oval 

19.1 Light: Notes on Philippine Termites, II 51 

with exception of apical segment which is narrower and 
elongated oval. Antennse and palpi very hairy, head with 
a few scattered hairs. Pronotum somewhat arched transversely 
and longitudinally, narrower than head, nearly twice as broad 
as long, anterior margin weakly concave, upturned, sides up- 
turned, rounded, receding to form rounded posterior margin. 
Mesonotum and metanotum much narrower than pronotum, 
sharply rounded posteriorly. 

Wings hyaline, veins difficult to make out, with exception of 
costa and radial sector which are large and a yellowish gray in 
color and run close together and parallel to one another to near 
the tip where they become much narrowed, lose their color, the 
radial sector soon uniting with costa; costa and radial sector 
joined in distal third of wing by 8 or 9 short, thick, cross veins as 
in P. inopinatus. Median of anterior wings very variable, united 
with cubitus through greater or less portion of wing; in many 
wings separating and uniting once or twice to form enclosed 
cells (see Plate 3, fig. 1) ; in other wings median arises from 
cubitus near distal third of wing as in P. flaviis; median or 
median and cubitus when united joined to radius sector by 
numerous, rarely branched, cross veins ; cubitus giving oif numer- 
ous (12 to 18) branched or unbranched veins to the posterior 
margin, which are united by numerous cross branches result- 
ing in a characteristic reticulation. Median of hind wing often 
arising from radial sector near base of wing or by branches 
from radial sector and cubitus ai)d united by numerous cross 
veins to radial sector and cubitus. All veins marked externally 
by lines of little hairlike rugosities. Anterior wing scales much 
larger than posterior pair; both light brown in color with ex- 
ception of oblique white line and bearing a few scattered 
spinelike hairs and a line of similar hairs along anterior border. 

Large soldier. — Head outstretched, body long and slender, 
body with head and mandibles as long as adult, head yellow, man- 
dibles reddish black, thorax and abdomen light yellow, hairs 
scattered on head, numerous on all other parts. Head con- 
siderably longer than broad, posterior border straight, corners 
rounded, broadest near posterior end, converging anteriorly, head 
with mandibles 2.75 to 3.00 millimeters long, without mandibles 
1.80 to 1.90 millimeters long, maximum width 1.40 to 1.65 milli- 
meters, minimum vndth 1.00 millimeter; head very low, flat- 
tened above; ventral gense somewhat arched; gula much nar- 
rowed at middle, nearly as broad posteriorly as in region of 

52 The Philippine Journal of Science 1921 

articulation of maxillae, twice as wide as at narrowest region; 
maxillary palpi longer than mandibles. 

Fontanelle circular, aperture directed somewhat posteriorly, 
channel of about same diameter running forward to base of post- 
clypeus. Antennal carinse prominent, edged with red, project- 
ing laterally over bases of antennse, ending at each posterolateral 
corner of postclypeus in a little rounded chitinous projection, the 
medial articulation of the mandible; postclypeus short and 
narrow, more than twice as broad as long ; anteclypeus very short 
and white; labrum short, tongue-shaped, with roundly pointed 
apex bearing two hairs. Mandibles as in P, flavus but with 
more gradually incurved tip; antennse of 18 or 20 segments, 
much like those of adult. Compound eye distinct, hyaline, 
lying in midlateral line of head just behind and considerably 
below posterior end of antennal carina. Pronotum considerably 
narrower than the head, somewhat arched with a middorsal longi- 
tudinal groove, slightly concave anteriorly, with shortly rounded 
anterolateral corners; sides rounding broadly into the nearly 
straight posterior margin which is very slightly concave in its 
central region; pronotum 1.35 millimeters broad and 0.65 milli- 
meter long. Pronotum broadest near anterior end, mesonotum 
near middle, and metanotum near posterior end; mesonotum 
narrower, shorter, and less heavily chitinized than pronotum. 

Small soldier, — Similar to large soldier but smaller and body 
broader and flatter with head carried at an angle to body, broader 
in proportion to length and not converging anteriorly so much 
as in large soldier. 

Measurements of Prorhinotermes luzonensis sp, nov., small soldier, 


Body length 6.00 

Head length, with mandibles 2.50 

Head lengfth, without mandibles 1.60 
Head width: 

Maximum I.45 

Minimum 1.00 

Pronotum width 1.25 

Pronotum length 0.55 

Worker. — Body much like large soldier. Head like adult 
but much less heavily chitinized and eyes much less prominent. 
Head sutures not very distinct and fontanelle much larger and 
roughly triangular; thickened, rounded articulations at lateral 
ends of clypeof rental suture distinct as in adult and soldier; 
antennse and mandibles same as in adult; eyes hyaline, larger 
than in soldier, and in same relative position. Antennse of 18 

1^' ^ Light: Notes on Philippine Termites, II 53 

segments. Pronotum 1 by 0.55 millimeter, mesonotum and meta- 
notum broader than pronotum. 

Nymph. — Very numerous nymphs with swollen floatlike wing 
pads united in midline and similar to those described by Snyder 
for P. simplex were present in the colony. These will be studied 
later in connection with the findings of Thompson and Snyder.^ 


As pointed out in the discussion under the genus, three collec- 
tions of Prorhinotermes have been made, all in Rizal within a 
kilometer or two of the Rizal-Bulacan boundary. The material 
on which this species is based came from a single colony found 
in a large hollow guava tree about a kilometer from the boun- 
dary. The colony was associated with a Hospitalitermes species, 
probably hospitalis (Haviland) or some nearly related species 
such as H. luzonensis (Oshima) . In tearing away the nest of 
the Hospitalitermes species a winged Prorhinotermes was seen 
but in the dusk was not at once recognized as a termite because of 
its light color, and its transparent wings and rounded body. 
Later, large numbers of all castes were collected, but unfor- 
tunately few data were obtained as to the relative positions and 
relations of the two forms. From the same place a number of 
specimens of Neotermes malatensis (Oshima) were obtained. 
These collections were made with Mr. R. C. McGregor on Octo- 
ber 4, 1920, and large numbers of winged adults and nearly 
mature nymphs were found in the nest.^ 

Several isolated winged adults have been collected. One 
(No. 197) was collected by me about the lights of the University 
Club, San Luis Street, Manila, September 27, 1920; another 
(No. 212) was collected from the lights in Quiapo, Manila, 
October 6, 1920; and two others (Nos. 66 and 245) I found 
in my former house in Ermita, Manila, one on August 28, and 
one on October 31, 1920. As adult specimens were found in 
large number in the colony (No. 205) on October 4, it seems 
probable that the winged adults take flight during August, Sep- 
tember, and October,'^ a few at a time probably, as the flying 

•Thompson, C. B., and Snyder, T. E., Biol. Bull. 36 (1919) 115. 

* Another trip to obtain more data about this colony showed the guava 
tree with the colony to have been destroyed, but another colony was found in 
another guava tree not far away (No. 436). Unfortunately it waa im- 
possible to make any extensive collections without destroying the tree. 

" It is interesting to note in this connection that winged adults of the 
same or a closely related species were taken about the lights in Cebu, in 
May, 1921. 

54 The Philippine Journal of Science 1921 

specimens collected were scattered individuals and I have seen 
no large flights. The finding of these adults in different parts 
of the city lends color to the belief that this species is more wide- 
spread than the limited number of colonies found would indicate. 
Both the house in Ermita and the University Club building are 
badly infested with termites, my house partly at least by Enter- 
mes (Microcerotermes) and the club house partly at least by 
Coptotermes but possibly also by Prorhinotermes. These ter- 
mites, being wood-dwellers and apparently building no covered 
galleries such as those of Eutermes (Microcerotermes) or 
Coptotermes and not dropping faecal pellets as do many species 
of the Kalotermitidse, are not easily located; hence the poverty 
of our collections.^^ 


The few species of this widely separated genus have not been 
studied as thoroughly as might be wished. Further study of a 
wide range of material may show that we have a single very 
variable species ranging from Formosa to Samoa of which P. 
japonictts, P. luzonensis, P. flavus, and P. inopinatus are merely 
variants, or subspecies. The great variation in the wing vena- 
tion of P. luzonensis would lend color to this belief. 

From other species as now known, P. luzonensis dif- 
fers in the following points, among others: From P. oceanicus 
Wasmann, P. krakataui Holmgren, and P. simplex Hagen in 
the greater size of the winged adult, and very strikingly from the 
soldier of P. krakataui in the greater number of antennal seg- 
ments ; from P. ivasmanni, a description of which I have failed 
to find, it probably differs in its larger size, since Holmgren 
believes P. wasmanni may represent the soldiers of P. ocean- 
icus; from each of the three closely related species P. japonicus, 
P. flavus, and P. inopinatus it differs in a number of minor 
points; from P. flavus, in wing venation, in toothing of man- 
dible of adult, and in relative breadth and length of pronotum; 
from P. japonicus and P. inopinatus, in greater convergence 
of anterior end of head of large soldier, etc. 

Prorhinotermes gracilis sp. nov. Text fig. 3. 

Types. — Dealated adult, large soldier, small soldier, and work- 
ers, No. 28 in type collection (from No. 150 in the general col- 
lection) . 

" Through the kindness of the Bureau of Public Works I am in position 
to report that this species has recently been found attacking the posts on 
the ground floor of the Bureau of Printing building. Further inspections of 
public buildings will probably show them to be quite common. 

19' 1 Light: Notes on Philippine Termites, II 55 

Cotypes. — No. 150 of general collection {McGregor and Light) , 
Rizal Province near Rizal-Bulacan boundary on Manila-North 
Road, September 29, 1920. 


Adult, — Same as P. luzonensis but darker brown in color. 
Wings not known. 

Soldier, — Like P. luzonensis but smaller, slenderer, lighter in 
color, head smaller, antennae of 15 or 16 segments, segments 
longer and slenderer than in P. luzonensis; compound eyes ves- 
tigial only, not protruding as in P. luzonensis; pronotum much 
smaller than in P, luzonensis. 

Worker.— Smaller y slenderer, lighter, antennae of 15 segments. 


Adult (dedlated), — Agrees very closely with P. luzonensis; 
color above generally darker brown, possibly due to greater age 
of this specimen which has made its flight while those of P. lu- 
zonensis were taken from the nest in the winged state. Ante- 
rior wing scales very broad, posterior ends overlapping in the 
midline (possibly an abnormality) . 

Large soldier, — In general like P, luzonensis but body smaller, 
about the size of the small soldier of P, luzonensis, slenderer, 
lighter in color, and less heavily chitinized; head smaller, 
pale yellow posteriorly but mandibles as dark as in P. luzo- 
nensis; mandibles as long as or longer than in P, luzonensis 
but slenderer; antennae with 16 segments, third obconic, usually 
distinctly larger and more heavily chitinized than second or more 
distal segments ; other segments longer and slenderer than in P. 
luzonensis ; compound eye vestigial, not protruding at all, repre- 
sented by a vague white area invisible except under microscope ; 
anteclypeus extremely short; narrowest part of gula more pos- 
terior than in P, luzonensis; pronotum smaller in proportion 
than in P, luzonensis. 

Measurements of Prorhinotermes gracilis sp. nov.^ 

, large soldier. 

Body length 



Head length, with mandibles 


Head length, without mandibles 


Head width: 





Pronotum width 


Pronotum length 


56 T'^^e Philippine Journal of Science 1921 

Small soldier. — Body short, fiat, darker yellow, and more 
heavily chitinized than large soldier; head larger than in large 
soldier, making an angle of about 45° with body. Pronotum 
slightly larger. Otherwise as in large soldier. 

Measurements of Prorhdnotermes gracilis sp. nov., small soldier. 


Body length 4.30-5.50 

Head, with mandibles 2.50-2.60 

Head, without mandibles 1.45-1.50 
Head width: 

Maximum 1.32-1.35 

Minimum 0.90-0.95 

Pronotum width 1.05 

Pronotum length 0.50 

Worker. — Long, slender, very lightly chitinized; thorax nar- 
row, head light yellow, body transparent white, abdomen colored 
dirty salmon to brown by intestinal contents. Head flattened; 
posterolateral region swollen, like P. luzonensis, but eyes only 
slightly developed, projecting very slightly. 


In view of the fact that this is the first case of two Prorhino- 
termes species found living together in the same region, and of 
the further fact that the two species were found in close prox- 
imity and have not been found elsewhere, it would seem, an 
obvious inference that we are dealing here with variational 
forms rather than with two distinct species. A consideration 
of the differences between the two species makes this position 
untenable, however. P. gracilis differs much more distinctly 
from P. luzonensis than the latter differs from P. japonicus, 
P. flavus, or P. inopinatus, with regard to which it is indeed 
a possibility that we are dealing with a very variable species of 
wide distribution. The lighter color and lack of chitinization 
and the very slight development of the compound eyes might 
be due to a more sheltered life habit ; the size difference might 
be a variation; but the very definite difference in number of 
antennal segments and their shape and size are difficult to ex- 
plain as mere variations. A more detailed study would show 
a host of minor differences and, unless the anterior wing scales 
of the one adult of P. gracilis are abnormal, the wings of the 
two species must differ very greatly. For these reasons I have 
felt it impossible to avoid making this a new species, to which 
I have given the specific name gracilis because of the slender 
form of the worker and large soldier. 

19, 1 

Light: Notes on Philippine Termites, II 




Collected by McGregor and Light on Septem- 
ber 29, 1920, from a hollow guava stub with 
living branches, near the Manila-North Road, in 
Rizal Province, about 100 meters from the Rizal- 
Bulacan boundary monument. 
This small colony was discovered 
while searching for Neotermes, 
one or more species of which 
are very common in guavas; in 
fact, a number of Neotermes 
specimens were collected at the 
same time, probably from tun- 
nels near the surface of the 
wood. In this connection it is 
interesting to note that P. luzon- 
ensis was also found in asso- 
ciation with, or at least in verj^ 
close proximity to, Neotermes. 
The P, gracilis colony was found 
living in a mass of wood pulp, 
probably fsecal matter, similar 
to that used by Neotermes to 
plug up points of entry into a 
limb, and deposited by them 
also in some of their workings. 
At the time this was thought to fig. 3. rrorhinoter 
be Neotermes waste, but I have 
not been able to verify this 
outline for contrast point and it secms probable that 

with antenna of P. . . , , , . , ^^ 

gracuis shown in it v^as produccd by the Pror- 
next figure. X 42. hiuotermes colony. In the cen- 
ter of this mass was a harder lump, apparently a royal chamber, 
in which was found a dealated adult male; the queen escaped 
or was overlooked. 

Genus LEUCOTERMES Silvestri sensu restricto 
Subgenus Leucotermes sensu stricto Holmgren. 


Adult,^^ — Yellow to brownish yellow; head oval; clypeus flat, 
short, and broad; labrum broad and convex; fontanelle small, 

Fig. 2. Prorhinoter- 
mes luzonensis sp. 
nov. Antenna in 


mea gracilia 
nov. Antenna in 
outline to contrast 
with that of P. 
luzonensis in pre- 
ceding fi R- u r (' . 

*^ From Banks and Holmgren. I have not seen the adult. 

58 The Philippine Journal of Science 1921 

dotlike, rather far back on head. Ocelli small or lacking. An- 
tennae of 15 to 17 segments, segments 2, 3, and 4 very short. 
Gula as long as broad. Pronotum flat, concave in front and 
behind. Anterior wing scale much larger than hind one. Wings 
slightly reticulate and strongly haired. Subcosta of anterior 
wing not extending beyond the wing scale. Radius running 
near anterior border with which it is often united. Radius 
sector simple, parallel to the anterior border to which it is often 
united apically by several small branches. Median usually 
simple and running nearer the cubitus than the radius sector. 
Cubitus with 8 to 12 branches to the hind margin. Radius 
of hind wing separate from anterior margin only within the 
wing scale. Tibiae with three apical spines. Cerci 2-segmented. 
"^ Soldier. — Head rectangular, with rather strongly inclined 
forehead clearly grooved in center. Clypeus short. Labrum 
rather long, tongue-shaped, with a sharp hyaline tip. Eyes 
lacking. Fontanelle in front of center of head. Mandibles 
with a large left and a small right basal tooth and beyond that 
with slight or no toothing. Pronotum flat, concave in front 
and behind. 

Worker, — Head rounded, oval, somewhat larger than the 
adult. Labrum large and broad. Head sutures not distinct. 
Fontanelle and plate present. Antennae with 13 to 15 segments. 

Leucotermes philippinensis sp. nov. Plate 4. 

Type. — No. 29 in type collection, soldier, worker, and nymph 
(from No. 128 of the general collection). 

Cotypes. — Nos. 128 and 132 (Miss Ursula B. Uichanco), Ma- 
nila, September 24 and 26, 1920 ; No. 432 (McGregor and Light) , 
Manila, March 25, 1921 ; No. 441 (McGregor) , Manila, April 3, 
1921, in the general collection. 


Adult. — Unknown. 

Soldier. — Labrum with pointed, awl-shaped, hyaline tip, fon- 
tanelle opening from distinct reddish brown tube; antennae of 
15 segments, the first very large and strongly swollen at distal 
end, all segments with a dense covering of short, distally directed, 
incurved hairs with scattered longer hairs; coxae and femora 
swollen, tibiae slightly so, tarsi very slender and more strongly 
chitinized. Body slender, particularly in region of thorax, 
covered with a coat of hair similar to that of the antennae, the 
short, incurved hairs directed posteriorly, hairing particularly 
heavy at posterior end of body. 

19, 1 Light: Notes on Philippine Termites, II 59 

Worker. — Antennse of 15 segments, first long and curved, not 
distally swollen; hairs as in soldier; body slender. 

Nymph. — Clypeus very greatly swollen; anterolateral corners 
of f rons raised, projecting ; antennae of 16 or 17 segments, haired 
as in soldier. 


Adult. — Unknown. 

Soldier. — Head, antennse, and maxillary palpi pale yellow ; head 
darker yellow anteriorly ; mandibles dark, transparent, brownish 
red ; bases of mandibles, internal mandibular articulations, mar- 
gins of antennal foveolse and **fontanelle tube" a light reddish 
brown; body white; anterior region of pronotum and the tarsi 
light yellow. Head long, parallel-sided, sides slowly rounding 
into rounded posterolateral corners; posterior margin straight; 
head dorsoventrally flattened with rounded sides (**thickly cylin- 
dricar' of Holmgren) . Forehead abrupt, making angle of about 
120 "" with the mandibles, laterally rimmed, centrally concave, 
dorsal rim concave; fontanelle, which is directed forward and 
distinctly visible internally as. a brown tube, lies at posterior 
end of concavity; head considerably thickened anteriorly, thick- 
est just behind level of antennae. Mandibles short (0.825 mil- 
limeter long), strong, nearly straight, tips somewhat incurved, 
right untoothed, left with roughened cutting edge and two small 
and low but distinct teeth close together near the base. Labrum 
0.45 millimeter long and 0.225 millimeter broad, tongue-shaped, 
ending in a very slender, awl-shaped, hyaline tip ; two large hairs 
at base of tip, beyond which the hyaline region has a length of 
0.09 millimeter. Anteclypeus very short, hyaline; medial ar- 
ticulations of mandibles at sides of postclypeus prominent and 
reddish in color. 

Antennse of 15 segments, relatively thick, much longer than 
mandibles, reaching about to posterior margin of head with scat- 
tered large hairs and very numerous, short, distally directed, 
incurved smaller hairs; basal segment very large, swollen dis- 
tally; second cylindrical, larger than third; others larger than 
second, suborbicular to broadly oval; apical segment slightly 
slenderer, long-oval; margins of antennal foveolse projecting, 
reddish brown. Pronotum narrower than head, more than half 
as long as broad, slightly concave anteriorly and less so posterior- 
ly, broadest near anterior end, lateral margins receding gradually 
and rounding posteriorly into straight posterior border, slightly 
emarginate in center; sides depressed, anterior margin very 
slightly upraised; divided by middorsal longitudinal groove; 

60 The Philippine Journal of Science 1921 

mesonotum much shorter and narrower, metanotum somewhat 
narrower ; lateral margins of mesonotum receding, those of meta- 
notum convex; posterior margin of each slightly emarginate in 
center. Coxae, femora, and, to some extent, tibise swollen ; tarsi 
very slender, strongly chitinized, yellow. Body slender, thorax 
particularly so, abdomen broadest near posterior end. Body 
and legs with scattered larger hairs and numerous smaller, in- 
curved, posteriorly directed hairs, similar to those of the 
antennae ; hairing most prominent toward posterior tip of abdo- 
men; head more sparsely haired. Cerci prominent. 

Measurements of Leucotermes philippinensis sp. nov., soldier. 


Body length 4.25 -4.75 

Head length 2.15 -2.25 

Head, without mandibles 1.10 -1.50 

Head width 0.80 -0.90 

Pronotum width 0.68 -0.78 

Pronotum length 0.425-0.50 

Worker, — Small ; body white ; head very pale yellow with excep- 
tion of exposed portion of mandible and mandibular articulations 
which are brownish yellow. Body, head, and legs covered with 
numerous hairs, the larger and more scattered hairs yellow, the 
shorter and more numerous, incurved, and posteriorly directed 
hairs white; body most heavily haired posteriorly, body longer 
than that of soldier ; head broadly oval ; clypeus swollen ; antennae 
of 15 segments, first long, curved, not distally swollen as in the 
soldier, apical segment slender, egg-shaped, longer than others 
with exception of first segment, haired as in soldier. Fonta- 
nelle and sutures not visible. 

Nymph. — Much like worker but abdomen very much longer 
and whiter, anterior wing pads 0.10 millimeter long; head much 
broader behind, clypeus very greatly swollen, distinctly bilobed ; 
anterolateral comers of frons high, projecting; compound eye 
projecting but little; antennae of 16 or 17 segments, shaped and 
haired like worker; fontanelle and Y-suture not visible. 


The specimens on which this species are based were collected 
for me by Miss Ursula B. Uichanco, head of the department of 
biology, of the Philippine Normal School. Three soldiers, a 
number of workers, and several nymphs (No. 128) were collected 
on September 24, 1920, from galleries in cracks in the cement 
floor and from the door jamb of the storeroom. One soldier 
and a few workers (No. 132) were collected on September 26, 

19.1 Light: Notes on Philippine Termites, II 61 

1920, from galleries on cement wall leading from a hole in the 
cement floor. On March 25, 1921, my attention was called by 
Mr. McGregor to termites building slender galleries on the 
cement supports of a porch of the Bureau of Science. These 
turned out to be this species, and two soldiers and numerous 
workers and nymphs were collected.^^ The latter showed no 
wing pads and it seems probable therefore that the period for 
emergence of the adult lies somewhere between September when 
the nymphs showed wing pads and March when they showed 
none. The species of Leucotermes have somewhat the same 
habits as Coptotermes but probably confine their attacks more 
completely to seasoned wood of buildings.^^ I hope in the future 
to make systematic collections in condemned buildings, which 
I surmise will disclose a considerably greater Prorhinotermes ^^ 
and Leucotermes population than is at present known. 


This species is very nearly related to Leucotermes indicola 
Wasmann. In the absence of comparative material it is impos- 
sible to be absolutely certain that we are not here dealing with 
a variety of that species ; but from all the data available on that 
species (Wasmann, Holmgren) I feel satisfied that, aside from 
minor differences, L. philippinensis differs from L. indicola 
Wasmann in the much greater length of the hyaline tip of the 
labrum of the soldier, in the greater length of the antennae and 
of the distal segments thereof, in the peculiar hairing of all 
castes, and in the presence of the definitely marked brown tube 
leading inward from the f ontanelle. As the species of this sub- 
genus are usually confined to a given region and are usually 
without coregional species, I have given the new species the name 

^^ On March 3, 1921, Mr. McGregor brought me specimens from a colony 
in a house in Paco, Manila, where repairs were being made due to termite 
damage. Examination showed them to be Leucotermes and a visit to the 
colony furnished a large series of this species. The nest, which superfi- 
cially resembled that of Coptotermes with its speckled yellow appearance, 
was much more compact with thicker walls and smaller and more rounded 
chambers. The nest was in the end of two floor sills. The wood attacked 
was not so thoroughly destroyed as in the case of attack by Coptotermes, 
particularly in the case of hardwood sills which were attacked mainly at the 
ends and on the sides. Pine pieces were entirely destroyed and replaced. 

^* More recently this species was found attacking growing sugar cane 
in an experimental plot near the Bureau of Science building. Mr. H. A. 
Lee, plant pathologist of the Bureau of Science, who brought this to my 
attention, tells me that he has frequent reports of termite damage to 
plants ! 


Plate 1 

Fig. 1. Kalotermes mcgregori sp. nov. Head and pronotum of short-headed 
soldier. Mandibles opened to show toothing. (Antennae imper- 
fect!) X 16.5. 
2. Kalotermes mcgregori sp. nov. Head and thorax, showing wing 
pads and antennae with 17 segments. X 16.5. 

Plate 2 

Fig. 1. Cryptotermes cynocephaliis sp. nov. Side view of head and part of 
thorax. X 30. 

2. CryptoterTYies cynocephalus sp. nov. Dorsal view of head and prono- 

tum. X 30. 

3. Planocryptotermes nocens sp. nov. Dorsal view of head and pro- 

notum. X 30. 

4. Planocryptotermes nocens sp. nov. Pronotum. X 30. 

Plate 3 

Fig. 1. Prorhinotermes luzonensis sp. nov. Head and pronotum of large 
soldier, x 19. 
2. Prorhinotermes luzonensis sp. nov. Winged adult. X 9. 

Plate 4 

Leucotermes philippi7iensis sp. nov. Dorsal view of soldier, x 24. 

Plate 5 

Fig. 1. A piece of picture molding attacked by Planocryptotermea nocens sp. 
nov., only a thin paperlike shell remaining. 

2. Cut end of above enlarged. 

3. Smaller pieces of pine eaten away, leaving extremely thin outer 


Plate 6 

A group of impressed faecal pellets of Planocryptotermes nocens sp. nov. 
enlarged about three times. 

TEXT figures 

Fig. 1. Kalotermes mcgregori sp. nov. Outline drawing af head and pro- 
notum of long-headed soldier. Note the absence of distinct 
eyespots and the presence of curious hyaline spots in anterolateral 

2. Prorhinotermes luzonensis sp. nov. Antenna in outline for contrast 

with antenna of P. gracilis shown in the next figure. X42. 

3. Prorhinotermes gracilis sp. nov. Antenna in outline to contrast v/ith 

that of P. luzonensis in the preceding figure, x 42. 








Light: Philippine Termites: II.] 

[Philip. Journ. Sci., 19, No. 1. 

Fios. 1 and 2. Cryptotermes cynooephalus sp. nov., head, side view and dorsal view, X 30. 3 
and 4. Planocryptotermes nocens sp. nov., head and pronotum, dorsal view, X 30. 


Light: Philippine Termites: II.] 

[Philip. Journ. Sci., 19, No. 1. 

Fig. 1. Head and pronotum of large soldier, X 19. Fig. 2. Winged adult, X 9. 

Light: Philippine Termites: 11.] 

[Philip. Journ. Sci., 19, No. 1. 


Light: Philii'PINE Tkrmitics : II,] 

[IHILIP. JOURN. Sci., 10, No. 1. 

F"iQ- 1. A piece of picture molding attacl<ed by Planocryptotermes nocens sp. nov., only a thin 
paperlike shell remaining. 2. Cut end of molding magnified. 3. Smaller pieces of pine eaten 
away, leaving extremely thin outer layer. 


Light: Philippine Termites: II. 1 

[Philip. Journ. Sci., 19, No. 1. 




By 0. Staff 
Of the Royal Botanic Garden^ Kew, England 

Closely allied to V, anceps Hk. f. (Cladium sinclairii Hk. f.) 
but differing in the 1- or 2-flowered, less compact spikelets and 
the wider, markedly white style-base of the ovary, slightly smaller 
nuts with a short pyramidal pubescent top, sharper and almost 
winged, broader, narrowing stipe. 

Perennial, 50 cm high. Culms compressed to ancipitous, with 
a single leaf from halfway up to the inflorescence. Basal leaves 
about 9, crowded, equidistant, sheathing portion about 4 to 5 cm 
long, blade 15 to 22 cm by 4 to 8 mm, acute, finely striate. 
Panicle over 20 cm long, formed of 5 or 6 downward distant 
anthelae; lowest anthela supported by a bract whose sheathing 
portion is compressed, sharply keeled, and 2.5 cm long, whilst 
the ensiform, shortly acute blade is 3 cm by 4 mm; bracts of the 
following anthelae rapidly decreasing in size, their blades nar- 
rowly linear to subulate; branches of the lowest anthela 2, of 
the following 5 or more and very unequal, the longest undivided 
for 6 cm or more; branchlets fascicled (up to 4 in a fascicle), 
unequal, the longest undivided to up to 1 cm, all bearing loosely 
approximate clusters of spikelets ; ultimate bracts ovate, cuspi- 
date, acuminate, bright chestnut-brown to fuscous, obscurely 
ciliolate. Spikelets broad-ovate, laterally more or less com- 
pressed, at length, more or less open with obliquely spreading 
glumes, about 3.5 mm long, 1- or 2-flowered, chestnut-brown to 
fuscous. Glumes ovate-oblong, acuminate or acute, cuspidate, 
2.5 mm long, very delicately scaberulous and ciliolate, the lower 
2 and the uppermost empty, the intermediate 1 or 2 fertile. 
Perianth none. Stamens 3, filaments much elongated at length, 
filiform, flexuous or in part twisted screw-fashion, remaining 
attached to the base of the nut. Ovary oblanceolate in outline, 
triquetrous, 1.2 mm long; style as long as the ovary, flliform 
from a thickened whitish pubescent base. Stigmas 3, slightly 

180365 o gg 

66 ^f^^ Philippine Journal of Science 

shorter than the style. Nut 1.6 mm long, pale brown in the 
center, whitish toward the ends, the top (style base) shortly 
pyramidal and pubescent, 0.4 mm long; stipe narrowly linear, 
0.4 mm long, its angles whitish, prominent acute. Grain 0.7 by 
0.5 mm. 

Luzon, Albay Province, Mount Mayon, Bur. Sci. 2935 Mearns. 


By A. H. Wells 
Chemist, Bureau of Science, Manila 

The results of investigations by Funk/ Eraser and Stanton,^ 
Chamberlain,^ Wilcox,* and Cooper^ have established the fact 
that beriberi is a deficiency disease. In the Philippine Islands, 
as in most tropical countries, the diet of the people is based 
upon rice. The investigations of the above-named authors, 
together with the findings of Braddon,^ Highet,^ Vedder,^ and 
others have sufficiently demonstrated that beriberi can result 
from a diet consisting of polished rice, and that extracts of 
these polishings contain neuritis-preventing substances. The 
work of these authors has been of the greatest value in the 
Philippine Islands, where beriberi is very prevalent and is 
traceable in many cases to the use of polished rice, both domestic 
and imported. 

In the preparation of rice for the market the glume or husk 
is first removed, then the grain is polished, and made white by 
the removal of the pericarp layer. This pericarp layer also 
constitutes the polishings, or tikitiki. 'In the Philippines tiki- 
tiki is sold as a cattle food, the best grades bringing about 4 
pesos (2 dollars) per sack of 50 kilograms. 

^Funk, Casimir, Journ. Phys. 43 (1911) 395; Journ. State Medicine 20 
(1912) 341; Trans. Soc. Trop. Med. 5 (1911) 86. 

^Fraser, Henry, and Stanton, A. T., Lancet 2 (1912) 1005; Lancet 1 
(1'909) 451; Lancet 2 (1910) 1755. 

"" Chamberlain, W. P., and Vedder, E. B., Philip. Journ. Sci. § B 6 (1911) 

* Wilcox, W. H., Brit. Med. Journ. No. 3081 (January 17, 1920) 73. 
"Cooper, E. A., Journ. Hyg. 12 (1912) 436. 

''Braddon, Leonard, Trans. Soc. Trop. Med. and Hyg. 11 (1909) 212; 
The cause and Prevention of Beriberi, London (1907). 

^Highet, H. C, Studies on Beri-Beri and its Prevention in Siam, Gov. 
of Siam, Bangkok, Siam (July, 1912). 

• Vedder, E. B., Philip. Journ. Sci. § B 7 (1912) 415. 


68 The Philippine Journal of Science 1921 

Regarding the antineuritic bodies existing in rice polishings, 
Vedder and Williams ^ give the following conclusions : 

(1) Undermined rice may be stored for one year in a damp place without 
losing its protective powers against polyneuritis gallinarimi. It is improb- 
able, therefore, that a rice which originally affords protection against beri- 
beri will lose this property by storage even in damp places. 

(2) The neuritis-preventing substances or vitamines contained in rice 
polishings are only slightly soluble in cold 95 per cent alcohol, since three 
successive extractions, using a total of six liters of alcohol to each kilo 
of polishings, fail to remove all of the neuritis-preventing substances from 
rice polishings. 

(3) Strong alkaline reagents such as sodium hydroxide, ammonia and 
barium hydroxide, destroy the neuritis-preventing vitamine, and the use of 
these reagents must be avoided in endeavoring to isolate this substance. 

(4) Basic lead acetate does not precipitate the neuritis-preventing vita- 
mine, and a considerable portion of this substance may be recovered from 
the filtrate. 

(5) The therapeutic properties of an alcoholic extract of rice polishings 
are greatly altered by hydrolysis (treatment with five per cent hydro- 
chloric or sulphuric acid). The unhydrolyzed extract is not poisonous, and 
is only slowly curative. The hydrolyzed extract is exceedingly poisonous 
in large doses and promptly curative in small doses. 

(6) We have confirmed Funk's observations by isolating a crystalline 
base from an extract of rice polishings by Funk's method. This base in 
doses of 30 milligrams promptly cured fowls suffering from polyneuritis 

(7) Funk's base or vitamine is present in rice polishings in considerable 
amounts, and only a very small portion of it can be obtained by Funk's 

(8) Two groups of substances (purine bases, choline like bases) may be 
isolated from rice polishings in addition to Funk's base, and are capable 
of partly or wholly protecting fowls fed on polished rice against polyneuritis 
gallinarum, but are incapable of curing fowls that have already developed 
the disease. The chemical nature of these two groups of bases requires 
further investigation. 

(9) We have confirmed the observation of Suzuki, Shimamura and 
Odake, that Funk's base may be precipitated from unhydrolyzed extract 
by tannic acid, but did not succeed in obtaining large amounts of this sub- 
stance by this method. 

(10) It is probable that this base or vitamine exists in food as a pyri- 
midine base combined as a constituent of nucleic acid, but that it is not 
present in the nucleins or nucleic acids that have been isolated by processes 
involving the use of alkalies, or heat. 

(11) The administration of unhydrolyzed extract of rice polishings to 
cases of adult wet beriberi, or to cases suffering from acute cardiac insuf- 
ficiency, results in the prompt dissipation of oedema, and relief of the 
cardiac symptoms. 

(12) The administration of unhydrolyzed extract of rice polishings to 

' Vedder, Edward B., Beriberi. New York, WiHiam Wood & Co. (1913) 
403, 404. 

19,1 Wells: Tikitiki Extract for Beriberi 69 

cases of dry beriberi is followed by little or no improvement in the para- 
lytic symptoms. 

(13) The administration of Funk's base to cases of dry beriberi is 
followed by an immediate improvement in the paralytic symptoms. This 
should remove the last doubt that dry beriberi is caused by the deficiency 
of this substance in the diet. It also finally proves that dry beriberi of 
man and polyneuritis gallinarum are essentially the same disease. 

(14) We have succeeded in curing a case of infantile beriberi (of the 
wet type) by administering that portion of the extract of rice polishings 
represented by the filtrate from the phosphotungstic precipitate. Since 
this filtrate does not contain Funk's base, this is evidence that wet beriberi 
is cured by some other substance. 

(15) Conclusions 11, 12, 13 and 14 are striking confirmatory evidence 
for the hypothesis previously stated by Vedder and Clark, that wet beri- 
beri and dry beriberi are two distinct conditions each being caused by the 
deficiency of a separate vitamine. 

Chamberlain and Vedder '^ obtained cures of infantile beri- 
beri by the use of extract of rice polishings made by extracting 
the fine powder with 90 or 95 per cent alcohol in the proportion 
of 3 liters of alcohol to each kilogram of polishings. Later Ved- 
der and Williams decided to extract the polishings three times 
with successive portions of fresh 95 per cent alcohol, using 3 
liters of alcohol to each kilogram of polishings for the first 
extraction and 1.5 liters of alcohol for each of the two following 
extractions. The extracts obtained were combined. Thus they 
observed a higher concentration of protective and curative sub- 
stances in the final extract. The same authors later decided to 
reduce the percentage of alcohol to 90 per cent strength and to 
heat the extraction to 60° or 70°C. The final method for the 
preparation of extract of rice polishings, originated by Vedder 
and Williams,^^ is as follows: 


Rice polishings or tiqui-tiqui may be obtained from any rice mill, but 
should preferably be from a recent milling. The finest grade of polishings 
should be carefully selected, since some of this product is very coarse 
and consists mostly of hulls. The tiqui-tiqui is first sifted to remove hulls 
and weevils. Gauze of about seven meshes to the centimeter i^ used for this 
purpose. This fine powder is weighed and mixed with 90 per cent alcohol 
in the proportion of three liters of alcohol to each kilo of polishings. It 
is then allowed to macerate for 24 hours. A glass jar or white enameled 
receptable [sic] serves for this pupose, and the mixture should be repeated- 
ly stirred or shaken, since the tiqui-tiqui sinks rapidly to the bottom, 
forming a densely packed mass which the alcohol penetrates with difficulty. 

'" Chamberlain, W. P., and Vedder, E. B., Bull. Manila Medical Society, 
No. 2, 4 (1912) 26. 

"Vedder, Edward B., Beriberi. New York, William Wood & Co. (191'^) 
405, 406. 

70 The Philippine Journal of Science 1921 

During the extraction the alcohol becomes of a deep green color, due to 
the fat that has been dissolved out. At the end of 24 hours the alcohol 
is siphoned off and filtered until absolutely clear. Since a very considerable 
quantity remains in the tiqui-tiqui, this should be squeezed in a press, or 
washed with fresh alcohol, and the residuum filtered and added to the 
alcoholic filtrate already obtained. The extraction should then be repeated 
several times, again using three liters of alcohol to each kilo of polishings. 
This is necessary because the neuritis-preventing substances are only 
slightly soluble in cold 90 per cent alcohol, and experience has shown that 
if the polishings are not repeatedly extracted the full therapeutic action 
of the polishings is not obtained. The combined alcoholic filtrate is then 
placed in a water bath provided with a thermometer, and an electric fan 
is so arranged as to throw a strong current of air on the surface of the 
alcohol. As a result of the heat and the movement of air the alcohol 
rapidly evaporates. It is essential that the temperature of the extract 
should not be permitted to rise above 80 °C., since extended observation 
has shown that greater heat is liable to decompose the active neuritis- 
preventing principle. Whenever the temperature of the extract approached 
[sic] 80 °C. the fire should be extinguished until the temperature drops. 
This process is continued until all the alcohol is evaporated. The residue is 
poured into a separating funnel and allowed to stand for about an hour, 
when it will be observed that the liquid has separated into two layers. 
The upper and larger portion is of a deep green color and consists of the 
fat. The lower ^nd smaller layer is brown in color, of syrupy consistency, 
and contains a number of substances that have been extracted by the 
alcohol. This lower layer is carefully drawn off, leaving the fat behind. 
It varies in amount, but about 25 cubic centimeters usually will be ob- 
tained from each kilo of polishings. The brown syrupy fluid so obtained 
from one kilo of polishings is diluted to 60 cubic centimeters with distilled 
water, whereupon a heavy precipitate is formed. This precipitate con- 
sists of substances that were soluble in alcohol, but are insoluble in water. 
After allowing the mixture to stand for a while the precipitate settles 
and the clear fluid is filtered off. This filtrate constitutes the extract 
as we have used it. Each 60 cubic centimeters contains the substances 
that have been extracted by this method from one kilo of polishings. 

In the latter part of 1913 the Bureau of Science commenced 
the manufacture of an extract of tikitiki for the cure of infantile 
beriberi. The method as outlined by Williams is as follows: 

Take 25 kilos (half of a sack) of tikitiki and soak in 75 liters 
of 20 to 25 per cent alcohol overnight or longer. Put in a 
cheesecloth sack and press slowly until pressure reaches about 
80,000 pounds. Obtain about 60 liters of extract, allow to stand 
and put in still and evaporate under 15 centimeters pressure 
or less. When concentrated to about 3 liters, remove, filter, 
and mix the clear liquid with an equal quantity of 90 per cent 
alcohol, which will cause precipitation. Let stand overnight, 
decant from the precipitate and evaporate under vacuum to about 
1.7 liters. Filter if not clear, and bottle. Sterilize the bottles 
at about 60° C, for twenty minutes, for three consecutive days. 

19,1 Wells: Tikitiki Extract for Beriberi 71 

One cubic centimeter of this extract equals 15 grams of tiki- 
tiki. The strength of the recovered alcohol will be from 20 
to 80 per cent by volume. This is diluted with water to make 
20 to 30 per cent alcohol and is used repeatedly. 


Ordinary dose: Three (3) teaspoonfuls daily. 

Serious cases: Double dose or more, according to requirement. 

This method was in use in the Bureau of Science until Jan- 
uary 1, 1916. The results obtained from the use of this product 
were so favorable that in 1916 it was decided to revise the method 
and to increase the production. During 1916 there was obtained 
a maximum possible production with the equipment at hand. 
Prior to this time one small copper still was utilized for con- 
centration. At this time a much larger still, which had been 
used for essential oil work, was put in operation for the first 
distillation. By the use of these two stills the production was 
brought to 6,687 bottles during 1916. During 1917 and 1918 
certain revisions in the methods of filtration made it possible 
to increase the production to 8,188 bottles in 1918. In 1919 
one small Elyria' glass-enameled still and a Sharpies laboratory 
centrifuge were installed. By the aid of these units larger 
quantities were worked up with a quicker and more effective 
separation of the inactive substance and a greater concentration 
for the final product. Year by year the demand for the product 
has increased until, at the present time, the Bureau of Science 
is unable to supply more than a small fraction of the actual 
requirements of the distributors. 

The finished product is placed in 50 cubic centimeter bottles, 
which are sealed, pasteurized, and labeled at the Bureau of 
Science prior to delivery to the various organizations for dis- 
tribution. Such organizations as the Public Welfare Board, La 
Liga Nacional para la Proteccion de la Primera Infancia, and 
the Philippine Health Service regulate the distribution, in order 
to insure that treatment shall be administered to patients by com- 
petent physicians. 


Rice polishings for use in this process must be free from 
insects, clean, and finely ground. The polishings from freshly 
milled rice of a new crop are preferable. The tikitiki (the 
rice polishings) is extracted for a period of forty-eight hours 
in a solution of alcohol of 25 per cent by weight (determined by 
use of the Abbe immersion ref ractometer) . The proportion of 

72 The Philippine Journal of Science 1921 

tikitiki to alcohol is one to two. Agitation is employed to a 
certain extent. After decantation, the residual sludge is passed 
to the press. The combined extracts are then passed to the 
distillation plant and evaporated under a pressure of 1 centi- 
meter with a maximum temperature of 75° C. The alcohol 
from this distillation is recovered and passed back to the 
extraction plant. This first distillation is stopped when the grav- 
ity of the extract has reached 1.18 at 70° C. The sirup at 
this density is allowed to stand overnight, or for a sufficient 
length of time for it to cool and settle completely. It is then 
decanted and the cloudy portion passed through a Sharpies 

The resulting clarified sirup is further freed of inactive 
substances by treating it with slightly over one-third of its vol- 
ume of 95 per cent ethyl alcohol. The gummy precipitate, well 
formed, is separated by means of the supercentrifuge and the 
alcoholic solution passed to the evaporator or smaller still, where 
the alcohol is recovered and the sirup concentrated under the 
same pressure and limit of temperature as before stated. The 
density of this final sirup is thus brought to 1.32. Upon 
cooling and standing overnight a flocculent precipitate of inactive 
susbtances forms and this is separated by again passing the 
sirup through the supercentrifuge. The finished product is 
then heated to 65 °C., bottled, pasteurized for three successive 
days at 62.5°C., labeled, lacquered, and delivered to distributors. 

Throughout the whole process as outlined above the extract 
comes in contact with metal only while in two of the stills that are 
tin lined. One still is glass enameled, although this is not abso- 
lutely essential. The precipitation, cooling, and storage are done 
in glass- or porcelain-lined vessels. 

Due to the tikitiki being somewhat heavy, agitation is advis- 
able in order to gain maximum extraction. Highly efficient 
recovery of alcohol is not obtained under the present method 
of coil and water condensation. 

By the method outlined a clear thick sirup of good flavor 
is obtained. One mil of this tikitiki extract represents the active 
constituents of 20 grams of tikitiki, or rice polishings. 

Tikitiki that has been in storage for a long time or that shows 
indications of mold growth has a tendency to produce an ex- 
tract which is high in acidity and which is not palatable. Also, 
an old tikitiki is usually highly infested with beetles and other 
insects. Such a product not only lacks the quality stated above 
but gives a much lower percentage yield of finished extract. 

19,1 Wells: Tikitiki Extract for Benheri 73 

Also, in the evaporation of extracts made from inferior grades 
of polishings, foaming takes place with consequent loss of time 
and yield. 

There are two grades of tikitiki ; that from the light-colored 
or white rice, and that from the dark or red rice. Exper- 
imentation with the tikitiki from the red rice did not give 
satisfactory results; the inactive substances were not easily 
precipitated nor wholly separable by centrifuge, and the 
extract obtained was of a very dark color and harsh in flavor. 
Further experiments will be made with the red polishings, and 
favorable results are expected. 

The process of the manufacture of tikitiki extract at the 
Bureau of Science has been well established. This extract is 
demonstrating by its therapeutic action that it possesses a high 
percentage of neuritis-preventing substances and that it is a 
cure for infantile beriberi. 

The Public Welfare Board at present (March, 1921) requires 
10,000 bottles of this extract monthly. The Philippine Health 
Service and other organizations are purchasing tikitiki extracts 
made by local druggists in order to fill their requirements. Such 
extracts are often made without the use of vacuum, and analysis 
by the Bureau of Science has shown that they contain glycerine, 
sugars, inactive substances, and in many cases high percentages 
of alcohol. Many of them give very little or no precipitate 
with phosphotungstic acid. The great number of these prep- 
arations made and disposed of on the local market may be 
taken as an indication of the prevalence of the disease. 

A plant with a capacity for the production of 15,000 bottles 
per month would permit the carrying out of a campaign for 
the treatment of beriberi throughout the Philippine Islands, 
and within one year from the time of installation of such a plant 
statistics on infant mortality would show a decided decrease. 

Tikitiki extract manufactured by the Bureau of Science from May 20, 19 lU 
to December 31, 1920, 

50 cc. 

May 20 to December 31, 19J4 1,161 

January 1 to December 31, 1915 3,997 

January 1 to December 31, 1916 6,687 

January 1 tio December 31, 1917 8,034 

January 1 to Decemb6r 31, 1918 &,188 

January 1 to December 31,, 1919 8,593 

January 1 to December 31, 1920 , 10,870 

Total ' 47,530 


Honorary Cv^todian of Hymenopteraf United States National Museum 

Subfamily SCOLIIN^^ 

The wasps of the subfamily Scoliinse can be easily distin- 
guished from their allies by the simple claws, the deeply emar- 
ginate inner margins of the eyes, the presence of only one 
calcarium on the intermediate tibia, the three spines on the 
male hypopygidium, the highly specialized tongue, and the 
general appearance. All of the species whose habits have been 
recorded prey on soil-inhabiting larvae of Coleoptera, and some 
of them (as, for example, Scolia manilae) have proven to be of 
great value in reducing these coleopterous pests. 
Key to Philippine genera of Scoliinse. 

1. Front wing with only two discoidal cells, the second recurrent being 

entirely wanting ,... Scolia Fabricius. 

Front wing with three discoidal cells, the second recurrent present.... 2. 

2. Second recurrent uniting with the first recurrent and not joining the 

cubitus Liacos Guerin. 

Second recurrent uniting with the second abscissa of the cubitus. 

Campsomeris Lepeletier. 

Genus SCOLIA Fabricius 

In the tabulating of the species of the genus Scolia I have ac- 
cepted the current method of dividing the genus into two sub- 
genera by the presence or absence of the second intercubitus (two 
or three closed cubital cells). This method of classifying the 
species often separates forms which are otherwise closely allied, 
and it seems that future work on the classification of these wasps 
will devise a more natural system by the use of body characters. 
The older writers paid but little attention to the structural de- 
tails; and without a large collection, which contains at least 
most of the described species, it will be impossible satisfactorily 
to arrange these insects by any system other than the one they 
adopted. The clypeus, antennae, pronotum, propodeum, first 
tergite, and pygidium offer useful characters, to say nothing of 
the shape of the head and the valuable suggestion of natural 
groups indicated by antigeny. 


76 The Philippine Journal of Science 1921 

Key to Philippine species of Scolia. 

1. Three closed cubital cells; subgenus Triscolia 2. 

Two closed cubital cells; subgenus Scolia 9. 

2. Thorax and abdomen "fusco-ferrugineis;" length, 11 millimeters. 

S. pseudoforaminata Gribodo. 
Thorax and abdomen (mostly at least) black 3. 

3. Apical abdominal segments clothed with reddish hair. 

S. rubiginosa Fabricius. 
Hair on apical segments black 4. 

4. Scutellum and metanotum marked with yellow; large species 5. 

Scutellum and metanotum black 6. 

5. Angles of pronotum and third tergite with a yellow spot. 

S. procer Illiger. 
Pronotum and abdomen black S. scutellaris Gribodo. 

6. Head marked with ferruginous or rufo-ferruginous 7. 

Head entirely black ^ 8. 

7. Antennae rufo-ferruginous; abdomen subcceruleous; less thait 20 milli- 

meters S. capitata Guerin. 

Antennae black; abdomen black; over 25 millimeters. 

S. philipplnensis sp. nov. 

8. Wings purplish S. bella sp. nov. 

Wings bronzy ;. S. bellina sp. nov. 

9. Head black, flagellum reddish S. auripennis Lepeletier. 

Head and antennae wholly or largely reddish 10. 

Head and antennae black 11. 

10. Propodeum shining, sparsely punctured; antennae entirely pale, those of 

the male shorter than the head and thorax S. erratica Smith. 

Propodeum with large close punctures on the dorsal, lateral, and pos- 
terior surfaces; antennae with scape black, those of the male fully 
as long as head and thorax S. westermanni Saussure. 

11. Small species, not more than 12 millimeters long.... S. manilae Ashmead. 

S. modesta Smith. 
Medium-sized species, over 18 millimeters long 12. 

12. Abdomen marked with ferruginous spots.. S. quadripustulata Fabricius. 
Abdomen entirely black 13. 

13. Wings violaceous 14. 

Wings brownish hyaline 15. 

14. Head large, posterior orbits wider than eye; propodeum black; female. 

S. megacephala sp. nov. 

Head not especially large, posterior orbits narrower; propodeum with 

a bluish reflection; female S. propodealis sp. nov. 

15. Hair of head and thorax above black; abdomen black; female. 

S. luzonensis sp. nov. 
Hair of head and thorax pale; abdomen with a bluish reflection; male. 

S. incerta sp. nov. 
Scolia (Triscolia) pseudoforaminata Gribodo. 

Triscolia pseudoforaminata Gribodo, Bull. Soc. Ent. Ital. 25 (1893) 

This species is not represented in the collection, but it should 
be easily recognized by its color which is described as "nigro- 
fuliginoso'' with thorax and abdomen as "fusco-ferrugineis.'' 

19, 1 Rohwer: Philippine Wasps 77 

Wings fuscous with obscure violaceous reflections. Length, 11 
millimeters. Only the male is known. 

Scoha (Triscolia) rubiginosa Fabricius. 

Although this species has been recorded from the Islands, it 
is not in the collections before me. The red hair on the apical 
segments should make it easily recognized. 

Scolia (Triscolia) procer Illiger. 

This species has been recorded from the Islands, but it is not 
represented in the collections I have seen from there. The typ- 
ical form has a yellow mark on the first tergite which a variety 
recorded by Gribodo (bimaculata Gribodo) lacks. In size and 
appearance this species is much like scutellaris. 

Scolia (Triscolia) scutellaris Gribodo. 

Triscolia scutellaris Gribodo, Bull. Soc. Ent. Ital. 25 (1893) 164. 
Scolia (Triscolia) whiteheadi Bingham, Ann. Mag. Nat. Hist. VI 16 
(1895) 441. 

There seems to be no good reason to doubt the above synonymy 
as both original descriptions apply well to the specimens listed 
below which are certainly all the same species. Besides the 
absence of yellow on the pronotum and abdomen the female is 
readily separated from procer by the more prominent tubercles 
on the pronotum. 

Luzon, Laguna Province, Mount Maquiling (Baker 2728) 
1 male: Manila, {George C. Lewis). Negros, Occidental Ne- 
gros, Bacolod, June 20, 1900 (A. T. Clifton) 1 female. Minda- 
nao, Iligan (Baker 3139), 

Scolia (Triscolia) capitata Guerin. 

This species closely resembles Scolia westermanni, which has 
only two cubital cells, but the scape and entire head are reddish 

Luzon, Manila (W. A. Stanton) 1 female: Bulacan Province, 
Baliuag (B. Arce) 1 male, under Bureau of Agriculture acces- 
sion No. 1591. 

Scolia (Triscolia) pMlippinensis sp. nov. 

Agrees very well with the description of Scolia alecto Smith 
except the abdomen is not iridescent and the propodeum is 
hardly emarginate posteriorly. The absence of yellow on the 
thorax and abdomen will separate it from procer Illiger. Ap- 
parently allied to S, intrudens Smith, but it cannot be assigned 
to that species because of the color of the wings. 

78 The Philippine Journal of Science 1921 

Female, — Length, 32 millimeters ; length of anterior wing, 30 
millimeters. Two strong diverging ridges between the anten- 
nae, the area between these ridges closely punctured and divided 
by a low carina; head smooth, with a few distinct punctures; 
antennae stout, the apical joint truncate; pronotum not swollen 
laterally, the surface with rather close punctures; mesoscutum 
smooth, with distinct punctures laterally; scutellum and meta- 
notum smooth, with scattered punctures on disk but with rather 
close punctures laterally; propodeum dorsally and caudally 
closely punctured, the posterior face slightly emarginate and dor- 
sally with an indistinct striation; first tergite with a median 
prominence basally; pygidium with distinct striae before tip. 
Black, clothed with long black hair which is especially dense on 
pronotum, sides of thorax, propodeum, and apical margins of 
abdominal segments; front, vertex, and posterior orbits to top 
of eye emargination ferruginous; wings purplish. 

Male. — Length, 28 millimeters; length of anterior wing, 24 
millimeters. Antennae slightly longer than head and thorax; 
hair on ferruginous part of head the color of the integument. 
Agrees with female except that it lacks the ridges between the 

Type locality. — Los Baiios, Laguna, Luzon. 

Type. — Catalogue No. 23582, United States National Museum. 

Luzon, Laguna Province, Los Baiios {Baker 1432) 1 female 
and 1 male; Mount Maquiling (Baker 3188) 1 male: Bataan 
Province, Lamao (C R. Jones) 1 female and 2 males, under 
Bureau of Agriculture accession No. 850. One of the last men- 
tioned males is slightly larger than the type. 

Scolia (Triscolia) bella sp. nov. 

Evidently allied to the Bomean S. crassiceps Cameron, but the 
frons is hardly "closely" punctured, the punctuation of the 
scutellum is different, and the mandibles are not fringed with 
rufous hair. Scolia nudata Smith is more coarsely sculptured. 
Scolia kollari Saussure is said to have the lateral dorsal area of 
the propodeum more sparsely sculptured than the median area. 
Scolia macrocephala Gribodo has the clypeus different and the 
anterior surface of the first tergite. impunctate. 

Fenmle. — Length, 17 millimeters; length of anterior wing, 13 
millimeters. Head large, as wide as thorax, posterior orbits 
wider than the greatest transverse diameter of eye; clypeus 
shining, with a few large punctures basally, convex medianly, 

19, 1 Rohwer: Philippine Wasps 79 

apical median portion depressed and broadly rounded; an oblique, 
low, rounded ridge over each antennal fossa; froiit convex, with 
large, well-separated punctures ; vertex and posterior orbits prac- 
tically impunctate; occiput with rather close, small punctures; 
ocelli rather small; postocellar line somewhat less than half as 
long as the ocellocular line; antennse stout, flagellar joints punc- 
tured, apical joint truncate, not as long as the two preceding; 
pronotum rounded, impunctate medianly, laterally uniformly 
rather closely punctured; scutum with sparse, large punctures 
except for two rather narrow, linear areas sublaterally ; scu- 
tellum, metanotum, and dorsal part of propodeum (except an 
impunctate area on inner basal part of each lateral area) with 
rather large uniform punctures ; tegulse, except at extremfe base, 
impunctate; lateral posterior areas of propodeum punctured, the 
median area impunctate; sides of propodeum sparsely punc- 
tured, separated from dorsal surface by a sharp carina; mes- 
episternum coarsely punctured; abdomen shining, very sparsely 
punctured ; first tergite short, the anterior surface with distinct 
but small punctures; apical tergites with large, rather close 
punctures; posterior calcaria stout, short, the longer one not 
half as long as basitarsus; second sternite truncate basally; 
pygidium rounded apically, shining, sparsely punctured; hypo- 
pygidium shining, lateral angles rather small, median lobe trape- 
zoidal in outline except for rounded apex. Black, basal tergites 
with a bluish reflection; hair rather sparse, black; wings 
uniformly deep violaceous. 

Male. — Length, 14 millimeters; length of anterior wing, 11 
millimeters. Head of normal size, subshining ; f rons with rather 
close, large punctures; vertex and occiput sparsely punctured; 
postocellar line but little shorter than the ocellocular line ; clypeus 
convex, punctured laterally, apical margin truncate; basal 
flagellar joints short (antennse wanting beyond fifth joint) ; 
scutum uniformly punctured; punctuation of scutellum, meta- 
notum, and propodeum like female but sparser; mesepisternum 
very coarsely punctured ; pygidium shining, sparsely punctured, 
obtusely angled apically; abdomen punctured like female. 
Colored like female except that in some lights there seems to 
be a sparse grayish pile on sides of thorax. 

Type locality. — Puerto Princesa, Palawan. 

Allotype locality. — Baguio, Luzon. 

Type. — Catalogue No. 23583, United States National Museum. 

80 The Philippine Journal of Science 1921 

Luzon, Mountain Province, Baguio (W. Robinson) 1 male, 
allotype. Panay, Antique Province, Culasi, June, 1918 (Mc- 
Gregor) 1 female, paratype. Palawan, Puerto Princesa 
(Baker) 1 female. 

Scolia (Triscolia) bellina sp. nov. 

Closely allied to Scolia bella, but smaller, the wings bronzy, 
the longer calcarium of hind tibia half as long as basitarsus, etc. 

Female, — Length, 13 millimeters; length of anterior wing, 10 
millimeters. Head large, as wide as thorax, posterior orbits as 
broad as greatest transverse diameter of eye, shining; frons 
convex and together with vertex and occiput with rather small, 
well-separated punctures; ocelli small; postocellar line half as 
long as the ocellocular line ; oblique ridges above antennse narrow ; 
clypeus shining, practically impunctate, convex medianly, the 
apical median margin depressed, and rounded; antennse short, 
shining, apical flagellar joints punctured, apical joint truncate, 
shorter than two preceding; pronotum rounded, impunctate 
medianly, uniformly punctured laterally ; tegulse, except basally, 
impunctate; scutum uniformly covered with rather large punc- 
tures; scutellum, metanotum, and dorsal part of propodeum 
(except impunctate inner half of lateral areas) with distinct, 
rather large punctures ; posterior face of propodeum punctured 
like the dorsal surface except more sparsely so ; sides of propo- 
deum with small, well-separated punctures, separated from the 
dorsal surface by a distinct carina, calcaria of hind tibiae stout, 
the longer one-half as long as basitarsus ; abdomen shining, very 
sparsely punctured, the apical segments more closely so; first 
tergite short, without carinse or tubercules, its anterior face with 
small distinct punctures; pygidium shining, sparsely punctured, 
apical margin rounded ; hypopygidium with small lateral spines, 
median lobe truncate. Black, basal tergites with a bluish reflec- 
tion ; hair sparse and black ; wings uniformly deep bronzy. 

Male. — Length, 11 millimeters; length of anterior wing, 9 
millimeters. Agrees closely v^ith the female. Head of normal 
size; ocellocular line one and one-half times as long as posto- 
cellar line; antennse shorter than head and thorax, the apical 
joint obliquely truncate and but little longer than the preceding; 
longer calcarium of hind tibia not half as long as basitarsus; 
pygidium and hypopygidium shining, sparsely punctured, their 
apical niargins obtusely angled. Color as in female, except that 
the legs and underparts of thorax have some of the hairs whitish. 

1^' ^ Rohwer: Philippine Wasps 81 

2'ype locality. — Davao, Mindanao. 

r^/^e.— Catalogue No. 23584, United States National Museum. 

Luzon, Tayabas Province, Malinao (Baker 5230) 1 male 
paratype. Mindanao, Davao {Baker 6921) 1 female, {Baker 
6907) 1 male. 

Scolia (Scolia) auripennis Lepeletier. 

Recorded from the Islands but not in the Philippine collection 
before me. 

Scolia (Scolia) erratica Smith. 

Scolia erratica Smith, Cat. Hym. Brit. Mus. 3 (1855) 88; Turner, 

Ann. Mag. Nat. Hist. VIII 8 (1911) 619. 
"Scolia molesta Saussure and Sichel, Cat. Spec. Gen. Scolia (1864) 


The specimens assigned to this species agree well with Saus- 
sure's account and are easily distinguished from Scolia wester- 
manni by the sparse punctuation of the propodeum and the 
shorter antenna of the male. 

Luzon, Laguna Province, Los Baiios {Baker 832) 1 female; 
Mount Maquiling, 1 male. Panay, Antique Province, Culasi, 
June 1, 1918 {McGregor) 1 male. 

Scolia (Scolia) westermanni Saussure. 

Scolia westermanni Saussure, Ann. Soc. Ent. France III 6 (1858) 

212; Turner, Ann. Mag. Nat. Hist. VIII 8 (1911) 619. 
Scolia erratica Saussure and Sichel, Cat. Spec. Gen. Scolia (1864) 


A single male collected by B. Arce and under accession 
No. 1591, Bureau of Agriculture, P. I. 

Scolia (Scolia) pianilae Ashmead. 

There is practically no doubt that this species was first 
described by Smith under the name Scolia modesta, but inas- 
much as none of the females before me have "a round macula" 
on the third segment (the mark when present being an oval 
or elongate spot) and none of the males agree with Smith's 
description in having the scutellum, metanotum, and propo- 
deum marked with yellow, it seems desirable to delay synony- 
mizing Ashmead's name until Smith's type can be examined. 

Most of the specimens of this common species show very 
little variation. One female from Los Banos, however, has the 
abdomen entirely without yellow marks, but two other females 
from the same locality (one under Baker No. 1709) have the 
marks of the abdomen reduced to a varying degree thus con- 

180365 6 

82 The Philippine Journal of Science 1921 

necting this single specimen with the others. The mark on 
the third tergite varies some in size and shape. Two males 
have the abdominal marks larger than usual but otherwise do 
not differ. One male, which may be somewhat immature, has 
the second and third tergites slightly reddish. 

Luzon, Mountain Province, Baguio (W, Robinson) (Baker) : 
Laguna Province, Los Banos (Baker 1708, 1709 , 1710, 18S6) : 
Union Province, Bauang (Baker i963) : Manila (W. A. 
Stanton). Negros, May, 1911 (C, V. Piper). Leyte, Tacloban 
(Baker). Panay, Antique Province, Culasi, June, 1918 
(McGregor). Mindanao, Davao (Baker 6906); Dapitan 

(Baker 319i). PALAWAN, Puerto Princesa (Baker). 


Scolia (Scolia) modesta Smith. 

See above remarks under Scolia manilae Ashmead. From 
the description the specimens described by Smith can be 
separated from manilae in the female by ''a round macula,'' on 
the third tergite, and in the male by the presence of yellow 
marks on the scutellum, metanotum, and propodeum. 

Scolia (Scolia) quadripustulata Fabricius. 

This species has been recorded from the Islands but is not 
in the material before me. The typical form has four ferru- 
ginous or yellowish marks on the abdomen; the wings are 
"nigro-chalybeis," and the head and thorax are almost impunc- 
tate. The hair is described as black. Some varieties with the 
abdomen entirely black have been described, and it is possible 
that the record of the species occurring in the Islands is based 
on an erroneous identification. Length, 15 to 20 millimeters. 

Scolia (Scolia) megacephala sp. nov. 

The large head allies this species with Scolia cephalotes 
Burmeister, but the black legs will readily distinguish it from 
that species. Disregarding the head it seems more closely 
allied to S. redtenbacheri Saussure. The large head seems to 
ally this species with the species of Triscolia with large heads. 

Female. — Length, 21 millimeters; length of anterior wing, 
17 millimeters. Head large, as wide as thorax, shining; vertex, 
occiput, and cheeks very sparsely punctured; posterior orbits 
somewhat broader than greatest transverse diameter of eye; 
clypeus shining, with large, separated punctures which are 
practically wanting on convex median portion, the apical margin 
depressed medianly and produced into a broad truncate lobe; 

19, 1 Rohwer: Philippine Wasps 83 

an oblique rounded ridge above and inside of each antennal 
fovea; front convex, with large sparse punctures; lateral ocelli 
small; ocellocular line one and one-half times as long as posto- 
cellar line; flagellum punctured, the apical joint truncate and 
not as long as two proceeding; pronotum rounded, uniformly 
punctured; tegulse with a few punctures; scutum shining, with 
large, separate punctures which are wanting in a rather small 
V-shaped area; scutellum and metanotum shining, uniformly 
punctured; propodeum emarginate posteriorly, dorsally with 
distinct punctures which are uniform except for polished, some- 
what triangular-shaped areas at base of lateral area; posterior 
face of propodeum sparsely punctured, the median portion with 
very few punctures; sides of propodeum with small separated 
punctures; mesepisternum coarsely reticulate-punctate; poste- 
rior calcaria simple, the longer one not quite half as long 
as basitarsus; abdomen shining, very sparsely punctured; first 
tergite subcampanulate, without tubercules or carinse, its ante- 
rior face rather closely punctured; second sternite subtuber- 
culate in basal middle; pygidium rather narrowly rounded 
apically, with rather close, irregular punctures; hypopygidium 
with lateral angles rather small and narrow, median lobe trape- 
zoidal in outline. Black, basal tergites with a distinct viola- 
ceous reflection; hair black; wings deep violaceous to apex. 

Type locality. — Mount Maquiling, Laguna, Luzon. 

Type. — Catalogue No. 23585, United States National Museum. 

Luzon, Mountain Province, Baguio (Baker) 1 female: La- 
guna Province, Mount Maquiling (Baker) 2 females (one the 
type). Panay, Antique Province, Culasi, May 13, 1918 
(McGregor) . 

Scolia (Scolia) propodealis sp. nov. 

Evidently allied to Scolia carbonaria Saussure and S. redten- 
bacheri Saussure but differs from the former in having both 
calcaria of hind tibiae simple and from the latter in larger size 
and more sparsely and finely punctured thorax. 

Female. — Length, 25 millimeters; length of anterior wing, 24 
millimeters. Clypeus flat, finely granular and with a few large 
punctures basally and an interrupted row a short distance before 
margin, the anterior margin broadly, roundly, slightly produced 
medianly; front with a transverse depression medianly, below 
which is a dorsoventral impressed line; head shining, not as 
wide as thorax, posterior orbits not swollen, as broad as greatest 

84 The Philippine Journal of Science 1921 

transverse diameter of eye; front and vertex with a few large 
punctures ; postocellar line about half as long as the ocellocular 
line; antennae stout, the apical joint truncate and as long as 
two preceding; pronotum rounded and with uniform punctures 
which are separated by a distance somewhat greater than their 
diameter; tegulse with quite a few punctures; scutum polished 
medianly; laterally with well-separated punctures; scutellum 
and metanotum with rather close punctures ; propodeum shining, 
deeply emarginate posteriorly, the central area dorsally and 
posteriorly and the sides of the lateral areas dorsally with rather 
small, well-separated punctures; sides of thorax shining, mes- 
episternum punctured like dorsal part of propodeum; posterior 
calcaria simple, the longer one somewhat more than half as 
long as basitarsus; first tergite without a tubercule or carina, 
campanulate, more closely punctured at base above; abdomen 
shining, very sparsely punctured; pygidium rounded, coarsely, 
irregularly punctured; hypopygidium sharply angled laterally, 
medianly produced into a broad, rounded, nearly parallel-sided 
lobe. Black, propodeum and basal tergites with a slight viola- 
ceous reflection; hair black; wings violaceous, beyond the 
venation more brownish. 

Type locality. — Mount Maquiling, Laguna, Luzon. 

Type. — Catalogue No. 23596, United States National Museum. 

Described from one female received from C. F. Baker. 

Scolia (Scolia) luzonensis sp. nov. 

The paler wings and pale hair on the sides of the thorax 
readily separate this from other Philippine species. It is 
possibly allied to Scolia melanosoma Saussure. 

Female. — Length, 22 millimeters; length of anterior wing, 
17 millimeters. Head normal, not as wide as thorax and with 
narrow posterior orbits, closely punctured, except for a trans- 
verse impunctate area behind ocelli and the more widely 
separated punctures on the front; clypeus uniformly punctured, 
convex medianly, the apical margin depressed, rounded median- 
ly and with transverse aciculations ; a small shining tubercule 
between bases of antennae; a narrow, oblique flange above each 
antennal fovea; front with a low, incomplete, transverse ridge 
which medianly is impressed by a line so as to form a broad V ; 
area surrounding ocelli more sparsely punctured ; ocelli of equal 
size, the postocellar line less than half as long as the ocellocular 
line; antennae stout, the flagellum punctured, the apical joint 
truncate and not quite as long as the two preceding; pronotum 

19, 1 Rohwer: Philippine Wasps 85 

rounded, closely covered with long, piceous hair which obscures 
the surface, the posterior margin with a fringe of short, yellowish 
hair; tegulse impunctate; scutum and scutellum closely punc- 
tured laterally, medianly impunctate; metanotum uniformly, 
rather sparsely punctured; propodeum nearly truncate, smooth 
shining, with a group of close punctures at base of median area 
and on lateral area just before middle; sides of propodeum 
nearly uniformly punctured, sharply separated from dorsal and 
posterior aspect ; posterior calcaria yellowish, simple, the longer 
one more than half as long as basitarsus; abdomen shining, 
sparsely punctured; first tergite short, without a tubercle or 
carina, its anterior surface sculptured like the dorsal; base of 
second sternite rounded; pygidium rounded apically, densely 
clothed with black bristles ; hypopygidium broadly rounded, the 
lateral spines sharp, but not conspicuous. Black ; tegulse yellow- 
ish; hair long, that on head (in part) , thorax above and abdomen 
(in part) black; hair of scape and a patch below antennal fovese 
slightly yellowish, that of back of head whitish; hairs of legs, 
sides of thorax, first tergite, apical margin of second segment, 
and apical margin of third sternite whitish ; tarsi piceous ; wings 
brownish hyaline, with a faint purplish sheen; venation brown. 

Type locality. — Baguio, Luzon. 

Type. — Catalogue No. 23587, United States National Museum. 

Described from one female received from W. Robinson. 

Scolia (Scolia) incerta sp. nov. 

This may be the male of Scolia luzonensis Rohwer. 

Male. — Length, 18 millimeters; length of anterior wing, 17 
millimeters; length of antennae, about 10 millimeters. Head of 
normal size ; clypeus convex, basally with a few large punctures, 
the anterior margin truncate ; area between the antennse closely 
punctured and with a small inconspicuous tubercle ; f rons shin- 
ing, sparsely punctured, with a transverse median depression 
which is angulate medianly to meet the narrow depressed area 
in front of anterior ocellus; vertex and occiput shining, prac- 
tically impunctate; postocellar line longer than the ocellocular 
line; antennse longer than head and thorax, scape shining, 
punctured; flagellum opaque, granular, the third joint slightly 
longer than fourth, apical joint slightly curved, obliquely trun- 
cate apically and somewhat longer than the preceding; thorax 
shining, sparsely punctured; calcaria yellowish, the longer one 
on the hind tibia more than half as long as basitarsus ; abdomen 
shining, sparsely punctured; first segment subcampanulate. 

86 The Philippine Journal of Science 1921 

longer than the second, pygidium and hypopygidium triangular 
apically. Black, thorax dorsally with a faint bluish reflection, 
basal tergites coeruleous; tegulae ferruginous; clypeus except 
margins, mandibles, line on pronotum laterally (interrupted in 
type and wanting in one paratype), anterior coxse beneath, 
femora, and the four anterior tibiae beneath yellowish; clothed 
with long whitish hair, sides, of thorax with whitish pile in 
addition; wings brownish hyaline with a slightly yellowish re- 
flection; Venation brown. 

In one paratype the lower inner eye margins are narrowly 

Type locality, — Mount Maquiling, Laguna, Luzon. 

Type, — Catalogue No. 23588, United States National Museum. 

Luzon, Laguna Province, Mount Maquiling {Baker 4962) 
1 male (type), (Baker 3190) 1 male: Bataan Province, Lamao 
(C. V. Piper). 

Genus LIACOS Guerin 

This genus, which in habitus closely resembles Scolia, is repre- 
sented by one species. 

Liacos (Triliacos) analis (Fabricius). 

A medium-sized black species with the two apical segments 
densely clothed with reddish hair. Wings blackish with a faint 
violaceous reflection basally. 

Luzon, Laguna Province, Los Baiios (Baker) 2 females; 
Mount Maquiling (Baker 15il) 2 males: Bataan Province, La- 
mao (C. V. Piper) 1 male, (C, R, Jones) 1 male, under Bureau 
of Agriculture accession No. 849. 

Genus CAMPSOMERIS Lepeletier 

All the species of this genus which occur in or have been 
reported from the Islands belong to the subgenus Campsomeris 
and have only two closed cubital cells. In this genus there is a 
marked antigeny, and therefore it is very difficult to associate 
the sexes. To add to this difficulty the species have not been 
fully described, thus always leaving some doubt as to the correct- 
ness of the identification. It seems to me that when more 
material from the Oriental Region has been studied additional 
species will have to be described and that some of the forms 
which are at present treated as unnamed varieties will be found 
to be sufficiently distinct to deserve names. 

The following key includes only species represented in the 
Philippine material before me, omitting the species Campso- 

19,1 Rohwer: Philippine Wasps 87 

meris grossa (Fabricius) and C. lindenii Lepeletier which have 
been reported from the Islands. 

Key to the Philippine species of Campsomeris, 

1. Females 2. 

Males 6. 

2. Abdomen, including hair, entirely black 3. 

Abdomen with at least pale hair bands 4. 

3. Body, including hair, entirely black C. reticulata (Cameron). 

Body black, hair of pronotum and back of head ferruginous; hair of 

front of head whitish C. aureicollis Lepeletier. 

. 4. Abdomen black, with whitish hair bands on each segment; hair of 
thorax white; wings hyaline, distinctly darker apically. 

C. annulata (Fabricius). 
Abdomen with apical margins of tergites yellow 5. 

5. Hair of head, thorax, and femora reddish C. aurulenta (Smith). 

Hair of head, thorax, and femora whitish C. aurulenta variety. 

6. Body entirely black 7. 

At least yellow markings on abdomen 8. 

7. Dorsal surface of propodeum with dense sericeous pile; abdomen dis- 

tinctly purplish; wings slightly paler C. luctuosa (Smith). 

Dorsal surface of propodeum without dense sericeous pile; abdomen 
bluish C. reticulata (Cameron) . 

8. Wings blackish; hair (not pile) of thorax black; yellow of abdomen 

reduced to elongate spots on second and third tergites. 

C. luctuosa (Smith). 

Wings hyaline to brownish; hair of thorax pale; yellow of abdomen 

forming bands 9. 

9. Markings yellowish; wings distinctly yellowish; pubescence of thorax 

yellowish C. aurulenta (Smith) . 

Markings whitish; wings dusky hyaline; pubescence of thorax gray.. 10. 

10. Abdominal bands broad, that of the second (also third) with a U- 

or V-shaped emargination medially; clypeus black, yellow at sides. 

C. annulata (Fabricius). 
Abdominal bands narrow, that on the second (and third less dis- 
tinctly) dilated at the sides; clypeus yellow, with a small black 
spot medially C. species? 

Campsomeris (Campsomeris) luctuosa (Smith). 

A male from Mindanao collected by Miss Ludlow and one 
from Mount Maquiling, Laguna, Luzon, collected by C. F. Baker. 
The specimen from Mindanao has narrow, elongate spots on 
sides of second and third tergites. 

Campsomeris (Campsomeris) reticulata (Cameron). 

Luzon, Mount Banahao {Baker 2727, 6910) males: Bataan 
Province, Lamao, 1911 (C. V. Piper) female. Panay, Antique 
Province, Culasi, June, 1918 (McGregor) female. Mindanao, 
Davao (Baker 6909) male: Misamis Province, Cagayan (Baker 
3796) male. 

88 The Philippine Journal of Science 1021 

CampsomeriB (Campsomeris) aureicoUis Lfepeletier. 

All of the Philippine specimens before me have the hair of 
the anterior part of the thorax reddish, and for the time being, 
at least, it seems advisable to use the name aureicoUis as a 
specific one. Ashmead in 1904 ^ added the name albicollis to 
the list of Philippine species, but as none of the forms have 
the hair of the collar pale the name albicollis should be removed 
until specimens of this variety are collected in the Islands. 

Luzon, Bulacan Province, Baliuag (B. Arce), Bureau of 
Agriculture accession No.* 1590: Laguna Province, Mount 
Maquiling (Baker) : Manila, July 14, 1901. Panay, Antique 
Province, Culasi, June, 1918 (McGregor). Mindanao, May, 
1911 (C. V. Piper) ; Dapitan (Baker 3187). All females. 

Campsomeris (Campsomeris) species? 

A number of males, which run to Campsomeris grossa in 
Bingham's key, are much smaller and represent a species which 
until further study cannot be identified. They may be the 
male of the female considered as a variety of aurulenta. 

Luzon, Mountain Province, Baguio (Baker) : Laguna Prov- 
ince, Los Banos (Baker) ; Mount Maquiling (Baker) ; Mount 
Banahao (Baker 4961). Mindanao, Davao (Baker 6908). All 

Campsomeris (Campsomeris) annulata (Fabricius). 

Luzon, Pangasinan Province, Resales (C. R. J ones) y Bureau 
of Agriculture accession Nos. 379, 380: Laguna Province, Los 
Banos (Baker 1431) male. Panay, Antique Province, Culasi, 
June, 1918 (McGregor). Leyte, Tacloban (Baker 3794) male. 
Mindanao, Iligan (Baker) female; Davao (Baker 6912) female, 
male; Cagayan (Baker 3795) male. 

Campsomeris (Campsomeris) aurulenta (Smith). 

This species is evidently allied to Campsomeris iris Lepele- 
tier where both the female and male run in Bingham's key to 
the species of British India. 

Luzon, Mountain Province, Baguio (W. Robinson, C. V. 
Piper) females: Bataan Province, Lamao (C. R. Jones), Bureau 
of Agriculture accession No. 1750 ; Laguna Province, Los Banos 
(Baker 522, 1834) female, male; Mount Maquiling (Baker 1835, 
1363) female, male; Mount Banahao (Baker) male. Pan AY, 

' Proc. U. S. Nat. Mus. 28 (1905) 152. 

19, 1 Rohwer: Philippine Wasps 89 

Antique Province, Culasi (McGregor) female, male. Minda- 
nao, Davao (Baker) male. 

Campsomeris (Campsomeris) aumlenta Smith variety. 

A single female from Baguio, Benguet (Baker) is smaller 
than the typical form and has the hair of the head, thorax, 
and femora v^hitish and much sparser. The abdominal bands 
are somewhat broader, that of the third segment being broadly 
produced medianly and with a wide angulate emargination. It 
may be that more material will make it desirable to name this 

Campsomeris (Campsomeris) lindenii Lepeletier. 

This species, which has usually been considered as a synonym 
of Campsomeris qicadrifasciata (Fabricius), has been reported 
from the Islands by Ashmead. It is not represented in any 
of the material I have seen from there, and the following char- 
acters are taken from the description by Bingham in the Fauna 
of British India: 

Female. — Length, 17 to 22 millimeters. Head and thorax 
sparsely punctured, disk of mesonotum smooth; black, pubes- 
cence fulvous to white; wings flavo-hyaline, with a fuscous 
spot at apex. 

Male. — Length, 17 to 22 millimeters. Black; apical margin 
of clypeus, posterior margin of pronotum, two spots on the 
scutellum, and transverse fasciae on the posterior margins of 
the first four or five tergites yellow; abdominal fasciae emar- 
ginate anteriorly; wings light flavo-hyaline. 

Campsomeris (Campsomeris) grossa Fabricius. 

This species has been recorded from the Islands but is not 
in the collections received from there. The female is 25 to 30 
millimeters long; black with the pubescence on the head and 
thorax fuscous, the abdomen with narrow, whitish hair bands. 
The wings are fusco-hyaline. The male is about 23 millimeters 
long, with broad transverse bands on the apical margins of 
the first four tergites yellow; the clypeus is black except for 
the narrow apical margin and a median spot which are yellow. 

Subfamily ELIDIN.^^ 

Only one representative of this subfamily is known from the 
Islands. The subfamily is readily separated from the Scoliin^ 
by the cleft claws; the entire (or nearly entire) eyes; the pres- 

90 The Philippine Journal of Science 

ence of a single spine, which curves dorsally, at the end of the 
male abdomen; the two calcaria on the intermediate tibia; the 
simple tongue; and the characteristic habitus. Many of the 
species of this subfamily lack the long hair that is character- 
istic of the Scoliinse. 

Elis (Mesa) tricolor longiceps Turner. 

The black thorax and abdomen, red head, and dark wings 
make it very easy to recognize this striking species. 

The females agree exactly with Turner's description of the 
subspecies, and the males differ from his description of Mesa 
crassepunctata as follows : Head distinctly narrowed behind the 
eyes; pronotum without obscure transverse striae, but polished 
and very sparsely punctured; first tergite longer than the 
second; abscissa of radius not one-fourth shorter than the 
third; first recurrent received by the second cubital just before 
the middle; length, 16 millimeters. The males may have been 
described by Bingham under the name Myzine dimidiaticornis, 
but the second abscissa of the radius is hardly as long as the 

Luzon, Mountain Province, Baguio (W. Robinson) 1 male, 
(Baker) 1 male: Laguna Province, Mount Maquiling (Baker) 
2 females. 


By Otto A. Reinking 

Plant Pathologist of the College of Agriculture and the Agricultural 
Experiment Station, Los Bancs 

Numbers I to IV of this series, on Higher Basidiomycetes from 
the Philippines and their hosts, show that a large amount of the 
destruction to forest trees and construction timber is due to 
fungi. Heretofore, the general belief has been that termites 
and other insects caused the greatest damage to woods in the 
Malay Archipelago. Fungi are widespread and are continually 
working under a variety of conditions. The destruction caused 
by them is greatest in damp situations and is particularly severe 
during the rainy season. The ravages of termites, on the other 
hand, occur only under certain conditions and in definite local- 
ities. According to the published lists, individual hosts may be 
attacked by at least ten different fungi. The damage done to 
certain structural timbers has been great. The proper kinds 
of wood to be used for building purposes will consequently 
depend upon the locality and the use to which they are to be put. 
As an example, guijo, Shorea guiso (Blanco) Blume, because it 
is attacked by a variety of fungi, should not be used for planking 
on a bridge which is exposed to rain and to high moisture condi- 
tions. Molave, Vitex parvifiora Juss., is much better for this 
purpose, as under the same conditions it is not so severely in- 
vaded by fungi. It seems probable, because of this fungus 
attack, that a preservative treatment of all woods used in ex- 
posed places would be beneficial. 

The following list of fungi is a continuation of the identi- 
fications of the higher Basidiomycetes collected on Mount Ma- 
quiling, in the vicinity of Los Banos, Laguna Province, Luzon, 
and in Mindanao. The collections have been made either by me 
or by students under my direction. I am indebted to Messrs. 
E. E. Schneider, J. M. Pascual, A. Barros, and L. Adona of the 
Bureau of Forestry for wood identifications. The majority of 
the determinations of fungi given in this list were made by 
C. G. Lloyd, of Cincinnati, Ohio. The species of fungi are 


92 The Philippine Journal of Science 1921 

grouped according to the classification of Engler and Prantl, 
with the host and the collector under each. The numbers refer 
to the College of Agriculture fungus herbarium. 



Annona muricata Linn., college ground, Los Banos, Collado 
5256, on dead branches. 

Annona reticulata Linn., college farm, Los Baiios, Reinking 
6Jt,00y on dead branches. 

Artocarpus sp.. Mount Maquiling, Soriano 5616, on dead 

Bixa orellana Linn., Mount Maquiling, Reinking 3928, on dead 

Clerodendron minahassae Teysm. et Binn., Mount Maquiling, 
Reinking i078, on dead branches. 

Ficus sp., Mount Maquiling, Paulino 5671, on decaying wood. 

Gliricidia sepium (Jacq.) Steud., Mount Maquiling, Reinking 
625i, on decaying wood. 

Hibiscus sp.. Mount Maquiling, Reinking Jlf297, on dead twigs. 

Jatropha curcas Linn., Mount Maquiling, Reinking ^283, on 
dead branches. 

Lansium domesticum Correa, Mount Maquiling, Sulit 5089, on 
dead branches. 

Leucaena glauca Benth., Mount Maquiling, Reinking 6U27, on 
dead branches. 

Mangifera indica Linn., college ground, Los Bafios, Hernandez 
957, on dead branches. 

Manihot utilissima Pohl, college ground, Los Baiios, Goco i072, 
on dead wood. 

Parameria sp., Mount Maquiling, Reinbing 3915, on dead 

Pterocarpvs indicus Willd., Mount Maquiling, Reinking 6^81, 
on dead branches. 

Streblus asper Lour., Mount Maquiling, Reinking H17, on 
dead twigs. 

Strijchnos nux-vomica Linn., Mount Maquiling, Reinking 
6M8, on dead branches. 

Triumfetta bartramia Linn., Mount Maquiling, Reinking 
3951, on dead branches. 

19, 1 Reinking: Basidiomycetes from the Philippines 93 


Prosopis vidaliana Naves, Mount Maquiling, Reinking 6U9U, 
on dead branches. 


Aglaia sp., Mount Maquiling, Reinking 3923, on dead 

Albizzia acle (Blanco) Merr., Mount Maquiling, Reinking 
i218, on dead twigs. 

Aleurites moluccana (Linn.) Willd., Mount Maquiling, Panga- 
niban 5130, on dead branches. 

Allaeanthus luzonicus (Blanco) F.-Vill., Mount Maquiling, 
Reinking 4^279, on dead branches. 

Annona muricata Linn., Los Baiios, Reinking 3779, on dead 

Annona reticulata Linn., college ground, Los Baiios, Reinking 
h232, on dead twigs. 

Clerodendron minahassae Teysm. et Binn., Mount Maquiling, 
Reinking i3i0, on dead twigs. 

Eriobotrya japonica Lindl., Mount Maquiling, Reinking 01 63, 
on dead branches. 

Erythrina fitsca Lour., college ground, Reinking H02, on dead 

Ficus hauili Blanco, Mount Maquiling, Reinking M25, on dead 

Jatropha curcas Linn., Mount Maquiling, Reinking W97, on 
dead wood. 

Melochia arborea Blanco, Mount Maquiling, Reinking U236, on 
dead twigs. 

Mussaenda philippica Rich, Mount Maquiling, Reinking H3S, 
on dead twigs. 

Psidium gimjava Linn., college ground, Los Baiios, Reinking 
U021, on dead wood. 

Pterocarpus echinatus Pers., Mount Maquiling, Reinking 
h292, on dead branches. 

Sapindus saponaria Blanco, Mount Maquiling, Reinking k2U7, 
on dead twigs. 

Solanum grandiflorum Ruiz et Pav., Mount Maquiling, Rein- 
king Jf322, on dead twigs. 

Solanum verbascifolium Linn., Los Baiios, Reinking 3766, on 
dead wood. 

94 The Philippine Journal of Science 1921 

Streblus asper Lour., Los Baiios, Reinking S772, on dead wood. 

Tecoma stans (Linn.) Juss., Mount Maquiling, Reinking 6ii>l, 
on dead branches. 

Theobroma cacao Linn., Los Banos, Reyes 8907^ on dead 

Trema amboinensis (Willd.) Blume, Mount Maquiling, Rein- 
king 4^098, on dead wood. 

UrcTia lobata Linn., Mount Maquiling, Reinking ^039, on dead 

Voacanga globosa (Blanco) Merr., Mount Maquiling, Reinking 
6A98, on dead branches. 


Anisoptera sp.. Mount Maquiling, Reinking SSIS, on dead 

Ficus sp.. Mount Maquiling, Baybay S^Oi, on dead wood. 


Annona sqvximosa Linn., college ground, Los Banos, Pangani- 
ban 5511, on dead branches. 

Leucaena glauca Benth., Mount Maquiling, Soriano 5625, on 
dead branches. 


TR EM ELLA Dillenius 

Probably Ficus sp.. Mount Maquiling, Cazenas 3307, on dead 


Caesalpinia pulcherrima (Linn.) Sw., Mount Maquiling, Rein^ 
king 6373, on dead branches. 

Shorea guiso (Blanco) Blume, Manila, Reinking 9701, on dead 


Mount Maquiling, Reinking 9803, on dead wood. 


Cassia siamea Lam., Davao, Ademesa 6059, on rotten trunk. 

Sapium merrillianum Pax et K. Hoffm., Mount Maquiling, 
Marilao 9720, on dead wood. 

Vitex parvi flora Juss., Los Baiios, Moncerate 6021, on railway 

19, 1 Reinking: Basidiomycetes from the Philippines 95 


Bamhusa sp., Mount Maquiling, Reinking 5069, on dead culms. 

Gliricidia sepium (Jacq.) Steud., Mount Maquiling, Reinking 
6253, on decaying wood. 

Leucaena glauca Benth., Mount Maquiling, Reinking 6266, on 
dead branches. 



Bamhusa sp., college campus, Los Banos, Marquez 3389, on 
dead culms. 

Gliricidia sepium (Jacq.) Steud., Los Banos, Fello 9716, on 
dead wood. 

Leucaena glauca Benth., Mount Maquiling, Nano 931^2, on dead 

Polyalthia sp.. Mount Maquiling, Serrano 9772, on dead wood. 

Shorea guiso (Blanco) Blume, Mount Maquiling, Ferrer 9765, 
on dead wood. 


Quercus sp.. Mount Maquiling, Marilao 963i, on dead wood. 

STEREUM Persoon 

Bambusa sp.. Mount Maquiling, Habaluyas 9620, on dead 


Mount Maquiling, Reyes 9656, on the ground. 

Mount Maquiling, Sison 9715, on the ground. 


Sapium merrillianum Pax et K. Hoffm., Mount Maquiling, 
Fello 9738, on dead wood. 


Probably Annonaceae, college campus, Los Banos, Pangani- 
ban 3356. 


Ficus sp.. Mount Maquiling, Panganiban 3379, on dead wood. 


Probably Leucaena glauca Benth., Mount Maquiling, Mendoza 
3396, on dead wood. 

96 The Philippine Journal of Science i»2i 


Los Baiios, Abisamis 9694^, on the ground. 


Probably Quercus sp., San Antonio, Los Baiios, Esguera 375, 
on dead wood. 


Ficus sp.. Mount Maquiling, Reyes 749, on dead wood. 


Ficus sp.. Mount Maquiling, Sanches 9749, on dead wood. 

SOLENIA Hoffman 

Ficv^ sp.. Mount Maquiling, Reinking 9784, on dead wood. 


Alstonia sp.. Mount Maquiling, Mendoza 8801, on dead wood. 


Ficus sp.. Mount Maquiling, Reinking 8880, on dead wood. 


Mount Maquiling, Obias 9687, on dead wood. 


Mount Maquiling, Sanches 9795, on soil. 


Bambusa sp., Los Banos, Habaluyas 9691, on dead culms. 


Mount Maquiling, Sanches 9798, on dead wood. 


Zizyphus sp.. Mount Maquiling, Fello 9787, on dead wood. 

^Incorrectly spelled as Cyphella fusco-disca Cooke in Reinking, Otto A., 
Higher Basidiomycetes from the Philippines and their hosts, II, Philip. 
Journ. Sci. 16 (1920) 170. 

19, 1 Reinking: Basidiomycetes from the Philippines 97 

IRPEX Fries 
IRPEX sp. 

Probably FicTis sp., college campus, Los Bafios, Reinking 3308, 
on dead wood. 


Bambusa vulgaris Schrad., college campus, Los Banos, Reyes 
4137, on dead culms. 

Leucaena glauca Benth., Mount Maquiling, Sarmiento 5174, on 
dead branches. 

GRAMMOTHELE Berkeley et Curtis. 

Ficus sp.. Mount Maquiling, Cazenas 991, on dead wood. 


Bambusa sp., Los Baiios, Habaluyas 9768, on dead culms. 

PORIA Per soon 
PORIA sp. 

Delonix regia Raf., Mount Maquiling, Nacion 9777, on dead 

Ficus sp.. Mount Maquiling, Ferrer 9642, on dead wood. 

Leucaena glauca Benth., Mount Maquiling, Caray 9631, on dead 

Mallotus sp.. Mount Maquiling, Abisamis 9728, on dead wood. 

Parkia timoriana (DC.) Merr., Mount Maquiling, Bacol 9800, 
on dead wood. 

Probably Shorea sp., college campus, Los Banos, Baybay 3342, 
on dead wood. 


Possibly Shorea sp.. Mount Maquiling, Sanches 9781, on dead 


Mallotus sp., Mount Maquiling, Rocafort 9751, .on dead wood. 


Mallotus sp.. Mount Maquiling, Ricafort 9i51, on dead wood. 

FOMES Fries 

Anisoptera sp., Kuruan, Zamboanga, Babao 450, on decaying 

180365 7 

98 The Philippine Journal of Science 1921 

Cocos nucifera Linn., Santa Cruz, Reyes 2955, on dead trunk. 

DipterocarpuSy Mount Banahao, Reinking il87, on decaying 

Leucaena glauca Benth., Mount Maquiling, Ferrer 9713, on 
dead branch. 

Shorea sp., Zamboanga, Tecson 456, on decaying wood. 

Tamarindiis indica Linn., San Antonio, Los Bafios, Reyes 4130, 
on dead trunk. 


Los Bafios, Abisamis 9711, on dead wood. 


Sandoricum koetjape (Burm. f.) Merr., Santa Cruz, Reyes 
5110, on dead stump. 


Mount Maquiling, Caray 9782, on dead wood. 


Mount Maquiling, Malabanan 9802, on dead wood. 


Mount Maquiling, Reyes 9683, on dead wood. 


Mount Maquiling, Marilao 9636, on dead wood. 


Gliricidia sepium (Jacq.) Steud., Los Bafios, Zabella 9712, on 
dead wood. 


Mount Maquiling, Ferrer 9767, on the ground. 


Possibly Euphorbiaceae, Mount Maquiling, Libunao 9615^ on 
dead wood. 
Mallotus sp., JMount Maquiling, Malabanan 9801, on dead wood. 


Probably Guttiferae, Mount Maquiling, Nantes 3395, on dead 


Mount Maquiling, Nacion 9669, on dead wood. 

19, 1 

Reinking: Basidiomycetes from the Philippines 99 


Gliricidia sepium (Jacq.) Steud., Los Baiios, Abisamis 9710, 
on dead wood. 


Alstonia sp., Mount Maquiling, Reinking 33 H, on dead wood. 


Mount Maquiling, Dadufalsa 9717, on dead wood. 


Zizyphus sp., Mount Maquiling, Collado 9682, on dead wood. 


Probably Alangium sp., Mount Maquiling, Malahanxin 9610, on 
dead wood. 


Probably Rubiaceae, Mount Maquiling, Reinking 3H2,^ on 
dead stem. 


Mount Maquiling, Bagui 5119, on the ground. 


Los Banos waterfalls, Los Baiios, Serrano 6169, on dead log. 


Annona squamosa Linn., Santa Cruz, Laguna, Reyes 6097, on 
dead bark. 

Bambusa sp., college campus, Los Baiios, Paulino 5669, on 
dead culms. 

Cordia myxa Linn., Mount Maquiling, Sarmiento 5656, on 
dead branches. 

Gliricidia sepium (Jacq.) Steud., college campus, Los Banos,, 
Bagui 6627, on old post. 

Mangifera indica Linn., Davao, Davao, Ademesa 6063, on 
rotten trunk. 

Shorea guiso (Blanco) Blume, Mount Maquiling, Ferrer 
9765, on dead wood. 

Vitex sp., Mount Maquiling, Reinking 6623, on dead branches. 


Mount Maquiling, Reinking 6087, on soil. 


Celtis sp.. Mount Maquiling, Nantes 661, on dead wood. 

100 The Philippine Journal of Science 1921 

Lagerstroemia speciosa (Linn.) Pers., Mount Maquiling, 
Manza 6598, on dead wood. 


Tabemaemontana pandacaqui Poir., Mount Maquiling, Rein^ 
king 6133, on dead branches. 


Possibly Burseraceae, Mount Maquiling, Malabanan 9698, on 
dead wood. 


Bambusa sp., Mount Maquiling, Bagui 5107, on dead stump. 

Bambitsa spinosa Roxb. (B. blumeana Schultes), Los Baiios, 
Ocfemia il23, on old bamboo posts. 

Ficus sp.. Mount Maquiling, Reinking 3U32, on dead wood. 

Leucaena glauca Benth., Mount Maquiling, Panganiban 5122, 
on dead stump. 

Mallotus sp., college campus, Los Banos, Navera 1065, on 
dead wood., college campus, Los Baiios, Divinagracia 3390, on 
dead stem. 

Polyalthia sp.. Mount Maquiling, Libunao 9769, on dead wood. 

Pterocarpus sp., Mount Maquiling, Aquino 6636, on dead 

Probably Rubiaceae, Mount Maquiling, Baybay 3372, on dead 


Mallotus sp.. Mount Maquiling, Reinking 9679, on dead wood. 


Probably Burseraceae, Mount Maquiling, Baybay 979, on dead 


Ficus sp., Mount Maquiling, Baybay 751, on dead wood. 
Pometia pinnata Forst., Mount Maquiling, Mendoza 3385, on 
dead wood. 


Mallotus sp.. Mount Maquiling, Nacion 9780, on dead wood. 


Acacia farnesiana (Linn.) Willd., Mount Maquiling, Pena 
5684', on dead branches. 

19,1 Reinking: Basidiomycetes from the Philippines 101 

Bauhinia tomentosa Linn., Mount Maquiling, Reinking US07, 
on dead twigs. 

Citrus sp., Lamao experiment station, Bataan, Reinking 5898. 

Cordia myxa Linn., Los Banos, Babao 9793, on dead wood. 

Euphorbiaceae, Mount Maquiling, Corrales 6617, on dead log. 

Gliricidia septum (Jacq.) Steud., Mount Maquiling, Reinking 
6248, on dead branches. 

Mangifera indica Linn., college ground, Los Banos, David 
6109, on dead branches. 

Tamarindu^ indica Linn., college ground, Los Baiios, Cuzner 
6i75, on dead log. 

Vitex sp.. Mount Maquiling, Piquing 6567, on dead branches. 


Koordersiodendron pinnatum (Blanco) Merr., Mount Maqui- 
ling, Nano 9599, on dead wood. 


Aleurites moluccana (Linn.) Willd., Mount Maquiling, Collado 
5219, on dead branches. 

Possibly Vitex sp., Mount Maquiling, Sison 9766, on dead 


Albizzia saponaria Blume, Cabantian, Davao, Ademesa 60^2, 
on dead trunk. 

Gliricidia sepium (Jacq.) Steud., Mount Maquiling, Collado 
5217, on dead branches. 

Mangifera indica Linn., Los Baiios, Reyes 3910, on dead 

Pithecolobium sp.. Mount Maquiling, Reinking 2970, on dead 

Tamarindus indica Linn., Mount Maquiling, Paulino 527i, on 
dead wood. 


Mount Maquiling, Malabanan 9626, on dead wood. 


Annonaceae, Los Banos, Piquing 5961, on railway ties. 

Bambusa sp.. Mount Maquiling, Lacson 5619, on dead culms. 

Bambusa spinosa Roxb. (B. blumeana Schultes), Anos, Los 
Baiios, Collado 5155, on dead culms. 

Cassia siamea Lam., Davao, Davao, Ademesa 60ii, on dead 

102 The Philippine Journal of Science 1921 

Diospyros discolor Willd., college campus, Los Banos, Rein- 
king il36, on old flagpole. 

Dipterocarpus sp., college campus, Los Banos, Morada 5928, 
on board. 

Myristica sp., Mount Maquiling, Mariano 97H, on dead wood. 

Neonauclea sp., Pangasinan, Soriano 6020, on railway ties. 

Parashorea plicata Brandis, college campus, Los Banos, 
Cuzner 3663, on old board. 

Pterocarpus sp., college campus, Los Banos, Panganiban 6650, 
on old board. 

Shorea philippinensis Brandis, Manila, Barros 10^99, on 


Ficus sp., college campus, Los Banos, Panganiban 3397, on 
dead wood. 


Areca catechu Linn., Los Banos, Reyes il35, on dead stem. 


Mount Maquiling, Malabanan 9635, on dead wood. 


Probably Dipterocarpus sp. or Anisoptera sp.. Mount Maqui- 
ling, Cazenas 610, on dead wood. 


Eucalyptus sp., Los Bafios, Minano 6013, on railway ties. 

Hopea sp., Los Banos, Ia7ig 6030, on railway ties. 

Intsia bijuga 0. Ktze., Los Banos, Reinking 5627, on railway 

Shorea guiso (Blanco) Blume, Los Baiios, Limbo 6008, on 
railway ties. 

Terminalia comintana (Blanco) Merr., Los Banos, Goco 5951, 
on railway ties. 

Vatica sp., Lo^; Baiios, Corrales 6027, on railway ties. 


Psidium guxijava Linn., college ground, Los Banos, Collado 
5Jf61, on decaying branch. 

Probably Shorea sp.. Mount Maquiling, Corcino 9794-, on dead 


Probably Celtis sp., college campus, Los Baiios, Reyes 2951, 
on dead wood. 

i^> ^ Reinking: Basidiomycetes from the Philippines 103 


Bambicsa sp., college campus, Los Banos, Reinking 3^.28, on 
dead culms. 

Cocos nucifera Linn., Los Banos, Reyes 3597, on dead trunk. 


Cordia myxa Linn., Mount Maquiling, Novey^o 6009, on dead 

hitsia bijuga 0. Ktze., Mount Maquiling, Catalan 5681, on 
decaying timber. 

Leucaena glauca Benth., Mount Maquiling, Reinking 3i43, on 
dead wood. 


Euphorbiaceae, Mount Maquiling, Corrales 6611^, on dead log. 

Gliricidia sepium (Jacq.) Steud., Los Banos, Reyes 516U, on 
dead trunk. 

Pterocarpus indicus Willd., Mount Maquiling, Sarmiento 5183, 
on dead log. 

Theobroma cacao Linn., Los Banos, Reyes Ui3, on dead 


Aleurites sp., Mount Maquiling, Reinking 9753, on dead wood. 


Bambusa sp.. Mount Maquiling, Bagni 5231, on dead stump. 

Lansium domesticum Correa, Los Banos, Reyes 3578, on 
decaying wood. 

Parashorea plicata Brandis, college campus, Los Baiios, 
Cuzner 366i, on old board. 

Terminalia comintana (Blanco) Merr., college campus, Los 
Banos, Collado 2996, on old board. 


Zea mays Linn., Los Banos, Ocfemia 3938, on dried ears. 


Mount Maquiling, Manza 607Jf, on dead branches. 


Acacia farnesiana (Linn.) Willd., Mount Maquiling, Man- 
gonon 5223, on dead branches. 

104 The Philippine Journal of Science 1921 

Bambicsa spinosa Roxb. (B. blumeana Schultes), college 
campus, Los Baiios, Reyes il38, on dead culms. 

Celtis sp., Mount Maquiling, Catalan 566^, on decayed log. 

Probably Euphorbiaceae, college campus, Los Banos, Reyes 
2987, on dead wood. 

Gliricidia sepium (Jacq.) Steud., Mount Maquiling, Sulit 
5617, on dead trunk. 

Leucaena glauca Benth., Mount Maquiling, Reinking 6265^ on 
dead branches. 

Parashorea plicata Brandis, Sirio, Lami, Zamboanga, Tecson 
i57, on decaying wood. 

Parkia timoriana (DC.) Merr., Mount Maquiling, Piquing 
5577, on dead log. 

Persea gratissima Gaertn., Mount Maquiling, Ocfemia 5523, 
on decaying branches. 

Pithecolobium sp., Mount Maquiling, Reyes 2972, on dead 

Prosopis vidaliana Naves, Mount Maquiling, Reinking 6^96, 
on dead branches. 

Tamarindus indica Linn., college campus, Los Baiios, Collado 
5 HO, on dead branches. 


Pterocarpus indicus Willd., college campus, Los Banos, Rein- 
king 6477, on painted board. 

Strombosia philippinensis Rolfe, Los Baiios, Aquino 6010, on 
railway ties. 


Probably Celtis sp.. Mount Maquiling, Santos 3i54', on dead 

Euphorbiaceae, Mount Maquiling, Corrales 6616, on dead log. 


Rapanea philippinensis (A. DC.) Mey., Mount Maquiling, 
Reinking 6513, on dead branches. 


Ficv^ sp.. Mount Maquiling, Catalan 5682, on living stump. 


Mount Maquiling, Reyes 9690, on dead wood. 

^^'1 Reinking: Basidiomycetes from the Philippines 105 


Alstonia sp., Mount Maquiling, Cazenas 3^28, on dead wood. 


Diplodiscvs paniculatits Turcz., Mount Maquiling, Nacion 
9665, on dead wood. 
Ophioliaceae, Mount Maquiling, Collado 9681, on dead wood. 


Mount Maquiling, Sarmiento 5622, on dead branches. 

CAM PAN ELLA P. Hennings 

Bamhusa sp., Los Banos, Reyes 9605, on dead culm. 

CANTHARELLA (Adans.) Linnaeus 

Ficus sp., Mount Maquiling, Habaluyas 9721, on dead wood. 


FiciLS sp., college campus, Los Baiios, Reinking 3^52, on dead 


Acacia farnesiana (Linn.) Willd., Mount Maquiling, Catalan 
5588, on dead branches. 

Aleurites moluccana (Linn.) Willd., Mount Maquiling, Pena 
5090, on dead branches. 

Anacardiaceae, college campus, Los Baiios, Pereira 5925, on 
post of house. 

Annona glabra Linn., college ground, Los Baiios, Corrales 
5i79, on dead wood. 

Bambusa sp., college campus, Los Baiios, Reinking 664,8, on 
dead culms. 

Bambusa spinosa Roxb. {B. blumeana Schultes), college 
ground, Los Baiios, Reinking 6078, on dead culms. 

Probably Burseraceae, college campus, Los Baiios, Reinking 
3^09, on dead wood. 

Calamus sp.. Mount Maquiling, Ocfemia 5678, on dead culms. 

Cassia sp.. Mount Maquiling, Collado U66, on dead roots. 

Celtis sp., Mount Maquiling, Catalan 566i, on decayed log. 

Cordia myxa Linn., Mount Maquiling, Salva Cruz 5191, on 
dead bark. 

106 The Philippine Journal of Science 1921 

Ficus sp., Mount Maquiling, Reinking 6125, on dead branches. 

Ficus nota (Blanco) Merr., Mount Maquiling, Manza 52A9, on 
dead branches. 

Intsia bijuga O. Ktze., Manila, Adona 10^97, on railway ties. 

Probably Koordersiodendron pinnatum (Blanco) Merr., col- 
lege campus, Los Baiios, Reyes 2935, on dead wood. 

Morns alba Linn., college farm, Los Banos, Palo 3813, on dead 

Persea gratissima Gaertn., Mount Maquiling, Ocfemia 5532, 
on dead branches. 

Pterocarpus indicus Willd., Manila, Adona 10495, on rail- 
way ties. 

Shorea mindanensis Desp., Manila, Barros 10498, on piling. 

Spondias lutea Linn., Mount Maquiling, Ocfemia 5476, on 
dead branches. 


Mijristica sp.. Mount Maquiling, Marilao 9752, on dead wood. 


Bambusa sp., Los Banos, Collado 9714, on dead stump. 


Barringtonia sp., Mount Maquiling, Abisamis 9677, on dead 


Mallotus sp.. Mount Maquiling, Corcino 9662, on dead wood. 
Pterocijmbium tinctorium (Blanco) Merr., Mount Maquiling, 
Dadufalsa 9725, on dead wood. 

PAN us Fries 

Mount Maquiling, Aquino 6632, on dead log. 


Polyscias nodosa (Blume) Seem., Los Banos, Abisamis 9770, 
on dead stem. 



Agathis alba (Lam.) Foxw., Mount Maquiling, Reinking 9700, 
on dead wood. 

Annonaceae, Mount Maquiling, Habaluyas 9622, on dead wood. 

19. 1 Reinking: Basidiomycetes from the Philippines 107 

Gliricidia sepium (Jacq.) Steud., Los Banos, Zabella 9756, on 
dead wood. 


Bambusa sp., Mount Maquiling, Reyes 96^6, on dead culm. 
Probably Polyalthia sp., college campus, Los Baiios, Reinking 
3H5y on dead wood. 



College campus, Los Banos, Reinking 6069, on nest of termites. 


Acacia farnesiana (Linn.) Willd. 

Lenzites repanda Pers., on dead branches. 

Polystictus (or Irpex) flavus Jungh., on dead branches. 

Schizophyllmn commune Fr., on dead branches. 
Agathis alba (Lam.) Foxw. 

Cyathus moiitagnei TuL, on dead wood. 
Aglaia sp. 

Auricularia cotmea Ehrenb., on dead branches. 

Alangium sp. 

Polyporus obovatus Jungh., on dead wood. 
Albizzia acle (Blanco) Merr., on dead twigs. 

Auricularia cornea Ehrenb., on dead twigs. 
Albizzia safonaria Blume. 

Polystictus occidentalis KL, on dead trunk. 
Aleurites sp. 

Trametes serpens Fr., on dead wood. 
Aleurites moluccana (Linn.) Willd. 

Auricularia cornea Ehrenb., on dead branches. 

Polystictus meyenii KL, on dead branches. 

Schizophyllum commune Fr., on dead branches. 
Allaeanthus luzonicus (Blanco) F.-Vill. 

Auricularia cornea Ehrenb., on dead branches. 
Alstonia sp. 

Favolus platyporus Berk., on dead wood. 

Pistillaria sp., on dead wood. 

Polyporus grammocephalus Berk., on dead wood. 
A nacardiaceae. 

Schizophyllum commune Fr., on post of house. 
Anisoftera sp. 

Auricularia moelleri Lloyd, on dead wood. 

Fomes applanatus Pers., on decaying wood. 

108 The Philippine Journal of Science 1921 

Annona glabra Linn. 

Schizophyllum commune Fr., on dead wood. 
Annona muricata Linn. 

Auricularia auricula-judae (Linn.) Schroet., on dead branches. 

Auricularia cornea Ehrenb., on dead twigs. 

Annona reticulata Linn. 

Auricularia auricula-judae (Linn.) Schroet., on dead branches. 
Auricularia cornea Ehrenb., on dead twigs. 

Annona squamosa Linn. 

Auricularia polytricha (Mont.) Sacc, on dead branches. 
Polyporus rigidus Lev., on dead bark. 


Cyathus montagnei Tul., on dead wood. 
Polystictus sanguineus Linn., on railway ties. 
Stereum involutum Kl., on dead wood. 

Areca catechu Linn. 

Polystictus spadiceus Bres., on dead stem. 

Artocarpus sp. 

Auricularia auricula-judae (Linn.) Schroet., on dead branches. 

Bambusa sp. 

Campanella cucullata Jungh., on dead cuhns. 

Corticium sp., on dead culms. 

Cyathus plicatulus Poep., on dead culms. 

Daedalea flavida Lev., on dead stump. 

Guepinia spathularia Schw., on dead culms. 

Lentinus badius Berk., on dead stump. 

Merulius consimilis Lloyd, on dead culms. 

Polyporus rigidus Lev., on dead culms. 

Polyporus zonalis Berk., on dead stump. 

Polystictus sanguineus Linn., on dead culms. 

Pterula taxiformis Mont., on dead culms. 

Schizophyllum commune Fr., on dead culms. 

Stereum sp., on dead culms. 

Trametes acuta Lev., on dead culms. 
Bambusa spinosa Roxb. (B. blumeana Schultes). 

Lenzites repanda Pers., on dead culms. 

Polyporus zonalis Berk., on old bamboo posts. 

Polystictus sanguineus Linn., on dead culms. 

Schizophyllum communemFr., on dead culms. 

Bambusa vulgaris Schrad. 

Irpex flavus KL, on dead culms. 

Barringtonia sp. 

Lentinus crinitus Swartz, on dead wood. 

Bauhinla. tomentosa Linn. 

Polystictus flavus Jungh., on dead twigs. 


Auricularia auricula-judae (Linn.) Schroet., on dead branches. 

19. 1 Reinking: Basidiomycetes from the Philippines 109 


Polypoms (Ganodermus) williamsianus Murr., on dead wood. 

Polystictus cervino-gilvus Jungh., on dead wood. 

Schizophyllum commune Fr., on dead wood. 
Caelsalpinia pulcherrima (Linn.) Sw. 

Tremella fuciformis Berk., on dead branches. 

Calamus sp. 

Schizophyllum commune Fr., on dead stems. 

Cassia sp. 

Schizophyllum commune Fr., on dead roots. 
Cassia siamea Lam. 

Guepinia fissa Berk., on rotten trunk. 

Polystictus sanguineus Linn., on dead branches. 
Celtis sp. 

Hexagona albida Berk., on dead wood. 

Lenzites repanda Pers., on decayed log. 

Polyporus semilaccatus Berk., on dead wood. 

Polystictis zelanicus Berk., on dead wood. 

Schizophyllum commune Fr., on decayed log. 
Citrus sp. 

Polystictus fiavus Jungh., on dead branches. 
Clerodendron minahassae Teysm. et Binn. 

Auricularia auricula-judae (Linn.) Schroet., on dead branches. 

Auricularia cornea Ehrenb., on dead twigs. 
Cocos NUCIFERA Linn. 

Fomes applanatus Pers., on dead trunk. 

Trametes acuta Lev., on dead trunk. 

Polyporus rigidus Lev., on dead branches. 

Polystictus fiavus Jungh., on dead wood. 

Trametes meyenii KL, on dead branches. 

Schizophyllum commune Fr., on dead bark. 
Delonix regia Raf. 

Poria sp., on dead wood. 

Polystictus sanguineus Linn., on old flagpole. 


Favolus spatulatus Jungh., on dead wood. 


Fomes applanatus Pers., on decaying wood. 

Polystictus sanguineus Linn., on board. 

Polystictus tabacinu^ Mont., on dead wood. 
Eriobotrya japonica Lindl. 

Auricularia cornea Ehrenb., on dead branches. 
Erythrina fusca Lour. 

Auricularia cornea Ehrenb., on dead twigs. 
Eucalyptus sp. 

Polystictus versatilis Berk., on railway ties. 

110 The Philippine Journal of Science 1921 


Hexagona albidcu Berk.j on dead log. 

Lenzites repanda Pers., on dead wood. 

Polyporus calignosus Berk., on dead wood. 

Polystictus flavus Jungh., on dead log. 

Trametes persoonii Mont., on dead log. 
P^ICUS sp. 

Auricularia auricula -jiidae (Linn.) Schroet., on decaying wood. 

Auricularia moelleri Lloyd, on dead wood. 

Cantharellus buccinalis Mont., on dead wood. 

Cantharellus infundibuliformis Berk., on dead wood. 

Cladoderris infundibuliformis Kl., on dead wood. 

Cyphella fulvodisca Cooke, on dead wood. 

Favolus albus Lloyd, on living stump. 

Grammothele 7nap2^a Berk., on dead wood. 
Irpex sp., on dead wood. 

Polyporus zonaljs Berk., on dead wood. 

Polystictus cryptomeniae P. Henn., on dead wood. 

Polystictus setulosus (P. Henn.) Lloyd, on dead wood. 

Poria sp., on dead wood. 

Pterula aciculae Lloyd, on dead wood. 

Schizophyllum commune Fr., on dead branches. 

Solenia globosa Lloyd, on dead wood. 

Stereum nigropus Lloyd, on dead wood. 

Tremella sp., on dead wood. 
Ficus HAuiLi Blanco. 

Auricularia cornea Ehrenb., on dead twigs. 
FiCUS NOTA (Blanco) Merr. 

Schizophyllum commune Fr., on dead branches. 
Gliricidia sepium (Jacq.) Steud. 

Auricularia auricula-judae (Linn.) Schroet., on decayed wood. 

Corticium sp., on dead wood. 

Cyathus montagnei TuL, on dead branches. 

Guepinia spathulata Schw., on decaying wood. 

Lenzites repanda Pers., on dead trunk. 

Polyporus annulatus Jungh., on dead wood. 

Polyporus gibbosus Nees, on dead wood. 

Polyporus rigidus Lev., on old post. 

Polystictus flavus Jungh., on dead branches. 

Polystictus occidentalis Kl., on dead branches. 

Trametes persoonii Mont., on dead trunk. 


Polyporus conchoides Mont., on dead wood. 
Hibiscus sp. 

Auricularia auricula- judae (Linn.) Schroet., on dead branches. 

HOPEA sp. 

Polystictus versatilis Berk., on railway ties. 


Polystictus versatilis Berk., on railway ties. 
Trametes meyenii Kl.^ on decaying timber. 

^^' 1 Reinking: Basidiomycetes from the Philippines m 


Auricularia auricula-judae (Linn.) Schroet., on dead branches. 
Auricularia cornea Ehrenb., on dead wood. 
Polystictus meleagris Berk., on dead wood. 
Schizohpyllum commune Fr., on dead wood. 
Lagerstroemia speciosa (Linn.) Pers. 

Polyporus semilaccatus Berk., on dead wood. 
Lansium domesticum Correa. 

Auricularia auricula-judae (Linn.) Schroet., on dead branches. 
Daedalea flavida Lev., on decaying wood. 
Leucaena glauca Benth. 

Auricularia auricula-judae (Linn.) Schroet., on dead branches. 
Auricularia polytricha (Mont.) Sacc, on dead branches. 
Corticiuyn sp., on dead wood. 
Fames applanatus Pers., on dead branches. 
Guepinia spathulata Schw., on dead branches. 
Irpex flavus KL, on dead branches. 
Lenzites repanda Pers., on dead branches. 
Polyporus zonalis Berk., on dead stump. 
Porta sp., on dead wood. 
Stereum ostreum Nees, on dead wood. 
Trametes meyenii Kl., on dead wood. 
Mallotus sp. 

Lentinus strigosus Schw., on dead wood. 
Polyporus zonalis Berk., on dead wood. 
Polystictus ajflnis Nees, on dead wood. 
Poria sp., on de^d wood. 
Poria fuligo Berk., on dead wood. 
Poria setulosa P. Henn., on dead wood. 
Mangifera indica Linn. 

Auricularia auricula-judae (Linn.) Schroet., on dead branches. 
Polyporus rigidus Lev., on rotten trunk. 
Polystictus flavus Jungh., on dead branches. 
Polystictus occidentalis Kl., on dead branches. 
Manihot utilissima Pohl. 

Auricularia auricula-judae (Linn.) Schroet., on dead branches. 
Melochia arborea Blanco. 

Auricularia cornea Ehrenb., on dead twigs. 

Schizophyllum commune Fr., on dead branches. 
Mussaenda philippica Rich. 

Auricularia cornea Ehrenb., on dead twigs. 
Myristica sp. 

Lentinus sp., on dead wood. 
Polystictus sanguineus Linn., on dead wood. 
Neonauclea sp. 

Polystictus sanguineus Linn., on railway ties. 

112 The Philippine Journal of Science 1921 


Favolus spathulatus Jungh., on dead wood. 

Polyporus zonalis Berk., on dead stem. 
Parameria sp. 

Auricularia auricula- judae (Linn.) Schroet., on dead branches. 
Parashorea plicata Brandis. 

Daedalea flavida Lev., on old board. 

Lenzites repanda Fers., on decayed wood. 

Polystictus sanguineus Linn., old board. 
Parkia timoriana (DC.) Merr. 

Lenzites repanda Fers., on dead log. 

Poria sp., on dead wood. 
Persea gratissima Gaertn. 

Lenzites repanda Fers., on decaying branches. 

Schizophyllum commune Fr., on dead branches. 


Lenzites repanda Pers., on dead wood. 
Polystictus occidentalis Kl., on dead wood. 


Corticium sp., on dead wood. 
Cyathus plicatulus Poep., on dead wood. 
Polyporus zonalis Berk., on dead wood. 
PoLYSCiAs NODOSA (Blume) Seem. 
Pleurotus sp., on dead stem. 


Polystictus cryptomeniae P. Henn., on dead wood. 
Prosopis vidaliana Naves. 

Auricularia brasiliensis Fr., on dead branches. 

Lenzites repanda Fers., on dead branches. 

Auricularia cornea Ehrenb., on dead wood. 

Polystictus xanthopus Fr., on decaying branches. 
Pterocarpus sp. 

Polyporus zonalis Berk., on dead branches. 

Polystictus sanguineus Linn., on old board. 
Pterocarpus echinatus Pers. 

Auricularia cornea Ehrenb., on dead branches. 
Pterocarpus indicus Willd. 

Auricularia auricula-judae (Linn.) Schroet., on dead branches. 

Lenzites striata Swartz, on painted board. 

Schizophyllum commune Fr., on railway ties. 

Trametes persoonii Mont., on dead log. 
Pterocymbium tinctorium (Blanco) Merr. 

Lentinus strigosus Schw., on dead wood. 


Hymenochaete rosea Lloyd, on dead wood. 
Stereum spectabile KL, on dead wood. 

19,1 Reinking: Basidiomycetes from the Philippines 113 

Rapanea philippinensis (A. DC.) Mey. 

Hexagona tenuis Hooker, on dead branches. 


Polyporus perversus Copel., on dead stem. 

Polyporus zonalis Berk., on dead stem. 
Sandoricum koetjape (Burm. f.) Merr. 

Fomes gibbosus Nees, on dead stump. 
Sapindus saponaria Blanco. 

Auricularia cornea Ehrenb., on dead twigs. 
Sapium merrillianum Pax et K. Hoffm. 

Guepinia fissa Berk., on dead wood. 

Stereum felloi Lloyd, on dead wood. 
Shorea sp. 

Fomes applanatus Pers., on decaying wood. 

Polystictus xanthopus Fr., on dead wood. 

Poria sp., on dead wood. 

Poria espimiltina Berk., on dead wood. 
Shorea guiso (Blanco) Blume. 

Corticium sp., on dead wood. 

Polyporus rigidus Lev., on dead wood. 

Polystictus versatilis Berk., on railway ties. 

Tremella fuciformis Berk., on dead wood. 
Shorea mindanensis Desp. 

SchizophylluTYi commune Fr., on piling. 
Shorea philippinensis Brandis. 

Polystictus sanguineus Linn., on piling. 


Auricularia cornea Ehrenb., on dead twigs. 


Auricularia cornea Ehrenb., on dead wood. 
Spondias lutea Linn. 

Schizophyllum commune Fr., on dead branches. 
Streblus asper Lour. 

Auricularia auricula-judae (Linn.) Schroet., on dead wood. 

Auricularia cornea Ehrenb., on dead wood. 
Strombosia philippinensis Rolfe. 

Lenzites striata Swartz, on railway ties. 
Strychnos nux-vomica Linn. 

Auricularia auricula-judae (Linn.) Schroet., on dead branches. 
Tabernaemontana pandacaqui Poir. 

Polyporus tabacinus Mont., on dead branches. 
Tamarindus indica Linn. 

Fomes applanatus Pers., on dead wood. 

Lenzites repanda Pers., on dead branches. 

Polystictus flavus Jungh., on dead log. 

Polystictus occidentalis Kl., on dead wood. 
Tecoma stans (Linn.) Juss. 

Auricularia cornea Ehrenb., on dead branches. 
180865 8 

114 The Philippine Journal of Science 

Tbrminalia comintana (Blanco) Merr. 

Daedalea flavida Lev., on old board. 

Polysticttcs versatilis Berk., on railway ties. 
Theobroma cacao Linn. 

Auricularia cornea Ehrenb., on dead branches. 

Trametes persoonii Mont., on dead branches. 
Trema amboinensis (Willd.) Blume. 

Auricularia cornea Ehrenb., on dead wood. 
Triumfetta bartramia Linn. 

Auricularia a/uricula-judae (Linn.) Schroet., on dead branches. 
Urena lobata Linn. 

Auricularia cornea Ehrenb., on dead branches. 
Vatica sp. 

Polystictus versatilis Berk., on railway ties. 
ViTEX sp. 

Polyporus rigidus Lev., on dead branches. 

Polystictus flavus Jungh., on dead branches. 

Polystictus meyenii KL, on dead wood. 


Guepinia fissa Berk., on railway ties. 
VoACANGA globosa (Blanco) Merr. 

Auricularia cornea Ehrenb., on dead branches. 
Zea mays Linn. 

Lenzites acuta Berk., on dried ears. 

Phlebia reflexa Berk., on dead wood. 

Polyporus mastoporus Lev., on dead wood. 


By F. T. McLean 

Professor of Botany y College of Agriculture, Los Banos 


The resistance of leaves to injection has been little investi- 
gated. In most of the physiological studies thus far reported, 
leaves have been 'treated mainly as instruments for gaseous 
interchange with the air and for photochemical reactions. A 
few ecological studies have been made of the relation of the 
form of leaves to the removal of water, and the opinion has 
been expressed that certain structures are beneficial in prevent- 
ing the clogging of the stomata with water and the flooding 
of the leaf tissues. So far as the writer is aware, there have 
been no studies of the pressures required to cause flooding. The 
resistance of leaves to the penetration of water may be important 
not only to prevent waterlogging of the tissues but also to pre- 
vent the entrance of certain leaf-disease organisms. Thus 
Pseudomonas citri Hasse, the cause of citrus canker, may be 
dependent mainly for its spread upon the permeability of the 
citrus leaves to water, as is suggested by McLean.^ He found 
that the differences in the structure of the stomata of a disease- 
resistant mandarin orange variety and a susceptible grapefruit 
were of such a character that water could enter the stomata of 
the former more easily than those of the latter. The structure 
and behavior of the canker bacteria are such that infection most 
probably takes place by means of continuous water columns, 
either through the stomata or through wounds. Whether in- 
fection will take place in this manner through the stomata or 
not depends upon the amount of pressure necessary to force 
water through the stomata. There appears to be need for more 
data on this particular point. 

There has been little study of the infiltration of the leaves 
of terrestrial plants with water, except in the case of a few 
epiphytic plants with special absorbing structures on the leaves. 
The infiltration of leaves with other liquids having a lower 

' McLean, Forman T., A study of the structure of the stomata of two 
species of Citrus in relation to citrus canker, Bull. Torr. Bot. Club 48 (1921) 


116 The Philippine Journal of Science 1921 

surface tension than water, such as alchohol, benzol, etc., was 
employed by Molisch ^ as an index of the openness of the stomata. 
He studied the rate of infiltration of these liquids without em- 
ploying pressure, and found that water was unsuitable for his 

The study here reported was undertaken to ascertain what 
pressures are necessary to cause infiltration of water through 
the stomata of Citrus leaves. A satisfactory method of causing 
water to enter the leaves through the stomata and of observ- 
ing the place and manner of entrance was devised and is here 
described. It was also found that penetration of water through 
the stomata can be easily induced under certain conditions, and 
that there are apparently differences in the amount of pressure 
required to cause penetration into the leaves of different varieties. 
Three varieties of Citrus were used in these tests: Washington 
navel orange, Szinkom mandarin orange, and Pernambuco grape- 


The equipment employed is shown in figure 1. It consists 
of a flat gas chamber g c of metal with a tube connection at 
each end. The bottom is of glass; the metal top has a circular 
aperture in it about 1.5 centimeters in diameter, over which the 
leaf I is placed. The gas chamber is mounted on the stage of a 
compound microscrope m, in position to examine the leaf sur- 
face under the low power (16-millimeter objective) . The right- 
hand tube of the gas chamber is connected to a mercury pressure 
gauge p, which is fitted with a scale s and mirror n to facilitate 
the rapid reading of the height of the mercury column by the ob- 
server seated at the microscope. The left-hand tube of the gas 
chamber is attached to an aspirator e, arranged to draw air 
from the gas chamber, the rate being controlled by means of the 
valve V. 


A piece of convenient size, usually about 3 centimeters in diam- 
eter, was cut from the leaf to be tested, and the upper epider- 
mis and palisade tissues were shaved off with a sharp razor 
from a small area in the center about 2 millimeters in diameter. 
Then the cut outer edges of the leaf were coated with paraffin 
(melting point, 45° C), and it was then sealed on to the aper- 
ture of the gas chamber with paraffin, the intact lower surface 

' Molisch, H., Opening and closure of stomata as shown by the method of 
infiltration, Zeitschr. Bot. 4 (1912) 106-122. 


McLean: Permeability of Citrus Leaves 



Fig. 1. Diagram of the apparatus used for measuring the pressures required to force water 
through the stomata of citrus leaves. 

118 The Philippine Journal of Science 1921 

of the leaf uppermost and the shaved thin portion in the center 
of the aperture. A glass ring was then fastened over the leaf 
with paraffin, and the small reservoir formed by the leaf and 
the glass ring was filled with tap water and covered with a mi- 
croscopic cover glass. The gas chamber was then connected with 
the pressure gauge and aspirator, and the leaf surface under 
water was examined with the low power of the microscope 
(16-millimeter objective) before any air was withdrawn from 
the gas chamber. In all cases the thinned portion of the leaf 
appeared opaque, due to the presence of air in the intercellular 
spaces, expect at the spots over oil glands, which were trans- 
lucent. The stomata had their outer chambers filled with large 
air bubbles. 

Air was then gradually withdrawn from the gas chamber un- 
der the leaf section by means of the aspirator. The partial vac- 
uum in the chamber and consequent pull on the water tending 
to draw it into the leaf tissues were indicated by the height of 
the mercury in the pressure gauge. The condition of the leaf 
was closely watched meanwhile, until the appearance of new 
translucent spots indicated that the leaf tissues were flooded, or 
until drops of water were observed by their shadows to be form- 
ing on the thinned portion of the leaf. Then the height of the 
mercury in the pressure gauge was recorded, the water reservoir 
was removed from the leaf, and the portions where water pene- 
tration had occurred were carefully examined. Actual breaks in 
the epidermis were observed in only two cases, and both of these 
were due to rapid and excessive application of pressure to the 
leaf. These two tests were valueless and were therefore dis- 
carded. In most cases it was clear, from the slow rate of infil- 
tration and from subsequent examination of the spot where the 
water had entered, that it had entered through the intact stomata. 
An effort was made to locate the particular stoma through 
which it first penetrated, \)\xi this could not be clearly distin- 
guished because of the low magnification, and in several cases 
the penetration first occurred through the thick tissues at the 
edge of the thinned portion, where the stomata could not be 
clearly seen. While examining the flooded portions of the leaf, 
a few measurements of the width of the ridge of entrance of 
the larger stomata in this portion were made, because the width 
of the stomatal aperture is thought to be the controlling factor 
in preventing the entrance of water. 

It was necessary to bring the leaves from the field to the labor- 
atory for testing, and it was found that the preliminary treat- 


McLean: Permeability of Citrus Leaves 


ment of the leaves had a great effect upon the injection pressure 
observed. Therefore, several different treatments were tried, 
and the results of these are presented in Tables 1, 2, 3, and 4. 
All of these tables are arranged in the same manner. An ar- 
bitrary number is given to each test for convenience in designa- 
tion. Pressure, causing penetration of water, was measured 
by means of the height, in centimeters, of the mercury column 
in the pressure tube. The averages of the width of the ridge 
of entrance of the stomata on the injected parts of the leaf are 
based on few measurements — usually ten and sometimes fewer — 
because the injected areas of the leaves were themselves small. 
The stomata varied considerably in size and only the larger ones 
were measured, as they were the ones through which the water 
would pass the most readily. 


The earlier tests were made on leaves which had been kept 
in the comparatively dim light of the laboratory for varying 
lengths of time before testing. The results of these tests are 
assembled in Tables 1 and 2, those of Table 1 including those 
which had been in the laboratory for one-half day or less, and 
those of Table 2 for a longer time. 
Table 1. — Tests of the pressure required to inject water into freshly 

gathered Citrus leaves, detached from twigs and kept in water for a 

short time. 




a. TO. p. TO. 

April 3- 


April 3 


April 3- 


April 4. 


April 4 


April 4 


April 4 


April 5 


April 5 


Kind of leaf. 

Young Pernambuco grape- 

Young Kishiu mandarin 

Young Szinkom mandarin 

Old Washington navel orange. 

Medium Washington navel 

Young Washington navel 

Half -grown Washington navel 

Medium Pernambuco grape- 
fruit.. _ _ 

Young Pernambuco grape- 


Width of ridge of 
entrance of stoma. 




cm. Hg. 































The Philippine Journal of Science 


Table 2. — Tests of pressure required to inject water into Citrus leaves , 
detached from twigs and kept with their petioles in water in the 
diffused light of the laboratory for one-half day longer. 


I No. 



a. w. 

p. m. 

April 4. 






April 4 _ 




April 5 




April 5 



Kind of leaf. 

Medium Washington navel orange. 
Half-grown Szinkom mandarin 


Old, yellow Szinkom mandarin 


Old Pernambuco grapefruit 

Young Szinkom mandarin orange. 
Young Pernambuco grapefruit... 


do . 


cm. Hg, 






Width of ridge of 
entrance of stoma. 








7.2 I 

A comparison of the leaves of different apparent ages of the 
three different varieties showed no clear relationship between 
age and injection pressure. The individual variations between 
tests were so great that, in order to eliminate one possible 
variant, subsequent tests were confined to comparatively young 
leaves. Therefore, no conclusion can be drawn from these obser- 
vations concerning the effect of age of the leaves upon the 
injection pressure. All of the above tests show exceedingly 
variable and generally high injection pressures, probably due to 
the unfavorable conditions to which they were subjected in the 

Several tests were made on leaves which had been immersed 
in water and placed in the window in bright diffuse light previous^ 
to testing. The results of these are assembled in Table 3. Some 
of these were tested before placing them in the window, and 
are thus entered in Table 2. These repeated tests on the same 
leaves are particularly valuable for comparison and to show the 
effect of conditions of exposure on the leaf properties. 

The injection pressures shown in Table 3, in which the leaves 
were immersed in water and kept in bright light before testing, 
are generally lower than those in Tables 1 and 2. Thus the 
values in Table 3 range from 2 and 5 millimeters in tests 23 and 
29 for Pernambuco grapefruit, to 15.5 and 18.0 centimeters 
in tests 27 and 28 for Szinkom mandarin. Further, a compari- 
son of the same leaf, tested in each of the two series, is illuminat- 

19, 1 

McLean: Permeability of Citrus Leaves 


Table 3. — Tests of pressure required to inject water into Citrus leaves, 
detached from the plant, immersed in water, and placed in bright light 
in the laboratory window. 






Apriie — 










Kind of leaf. 

Medium Pernambuco grapefruit... 


Width of ridge of 
entrance of stoma. 




p. m. 

cm. Hg. 








Young Pernambuco grapefruit 

Medium Pernambuco grapefruit ... 
Young Szinkom mandarin orange- 
Young Pernambuco grapefruit 


Young Szinkom mandarin orange. . 

do - 

do — 


Young Pernambuco grapefruit 






ing. Thus a medium Pernambuco grapefruit leaf in the first 
series (No. 8, Table 1) gave an injection pressure of 12.0 cen- 
timeters, while the same leaf, under the more favorable treat- 
ment in the second series (Nos. 18, 19, and 21, Table 3), gave 
values of 9.5, 2.0, and 3.0 centimeters successively. The lowest 
pressures obtained in this series, as shown in Table 3 for Per- 
nambuco, are of such magnitude as might easily occur in leaves 
on the trees, due to changes of temperature when the leaves 
are wet, or even to rapid expansions and contractions of the air 
chambers when the leaves are bent in being blown about by the 
wind. The combination of abundant moisture and light provided 
in these tests appears to have caused the stomata to open and 
make the entrance of water easy. 

The tests in Tables 1, 2, and 3 do not nearly approach the nor- 
mal conditions to which the leaves on the trees are subjected 
during bright sunny weather. Therefore, a third lot of leaves 
were left attached to twigs, which were put in water and placed 
so as to be fully exposed to the sun. These were tested at inter- 
vals during the day, and the results are tabulated in Table 4. 

The young leaves on twigs and exposed to full insolation show 
great variations in the pressures required to inject them with 
water. Using similar Szinkom leaves for all tests, the values 
varied from 5.5 to more than 38.0 centimeters. Since these 


The Philippine Journal of Science 


leaves were fully exposed to sunlight, these differences may have 
been caused by variations in the rates of transpiration. 

Table 4. — Tests of pressure required to inject water into Citrus leaves^ 
attached to twigs and standing in water, fully exposed to the sun before 
testing them. 



Kind of leaf. 

April 7 
April 7 
April 7 
April 7 
April 7 
April 7 

a. m. V' w. 


Young Szinkom mandarin orange. 
do - - 



Width of ridge of 
entrance of stoma. 



cm. Hg: 









38.0 + 



With the small number of data presented, and these of such a 
highly variable character, it is obviously unsafe to make any 
more than the most general sort of conclusions. The effects of 
the different treatments employed appear to be quite evident from 
the foregoing comparisons of results. Less satisfactory con- 
clusions can be reached regarding the differences between the 
different varieties, and the possible correlation between the Avidth 
of the ridge of entrance of the stoma and the injection pressure. 
Certain generalities are allowable from the data shown. Thus 
a consideration of the lowest values obtained, as shown in Table 
3, indicates that Pernambuco grapefruit is usually more easily 
injected than Szinkom mandarin, since of the five injection pres- 
sures below 5 centimeters four, including the two lowest values 
(2 and 5 millimeters), are for Pernambuco and only one (1.6 
millimeters) is for Szinkom. Further, the average injection 
pressure for Pernambuco is 7.9 centimeters, while for Szinkom 
(taken from the first three tables only, for those in Table 4 are 
not considered to be at all comparable to the others) it is 18.3 
centimeters. Likewise, the average width of the ridge of en- 
trance of the stomata of Pernambuco is 7.0 /* and of Szinkom, 
4.1 /i. Thus the average values for the two varieties indicate 
that Pernambuco grapefruit is more easily injected and has 
wider stomatal apertures than Szinkom mandarin. 

Another more extensive series of tests of the injection pres- 
sures of Citrus leaves is being made on leaves attached to the 
trees in the field, using a different technic. The results of these 

19. 1 McLean: Permeability of Citrus Leaves 123 

will be the subject of a later contribution. The data thus far 
obtained on the leaves on the trees bear out in a general way 
the tentative conclusions stated above. 


1. A method is described for determining the pressure required 
to force water through the stomata of leaves. 

2. The pressure required to inject leaves of the same variety 
seems to vary greatly in accordance with the treatment before 
testing, and seems to be lowest when the leaves are exposed to 
bright diffused light and well supplied with moisture. 

3. Apparently Szinkom mandarin orange leaves require on the 
average more than twice as much pressure to inject them with 
water as is required for the leaves of Pernambuco grapefruit. 

4. The conclusion stated in 3 seems to be correlated with the 
average width of the stomatal aperture through the epidermis, 
that of Szinkom mandarin orange being a little more than half 
as wide as that of Pernambuco grapefruit. 



Fig. 1. Diagram of the apparatus used for measuring the pressures required 
to force water through the stomata of citrus leaves. 



128 The Philippine Journal of Science 


The information most eagerly sought by those entering upon 
the work among the feeble-minded is naturally how to easily 
recognize the various forms of mental defect, in order that they 
may define, and meet promptly, the special needs of those with 
whom they are brought in daily contact. 

To this end, types of various grades are useful as sign-posts 
pointing the way to successful diagnosis of defect — ^mental, 
moral and physical. In defining types many points, such as 
have been indicated by tests, as well as by the stigmata of 
degeneration noted in the individual, are to be considered. 

Appended herewith will be found the educational classification, 
which, as the outgrowth of a close study of cases and careful 
adaptation to needs — indorsed by both physicians and teachers — ^ 

has proven in a long experience the best one as simplifying the 
tasks of all engaged in the work. 

This classification is arrived at by first separating broadly the 
untrainable idiot from the trainable imbecile in asylum, custo- 
dial, and school division; next by dividing the imbeciles into 
grades of mentality for the awakening and further development 
of power along lines suited to the capacity of each; and finally 
by indicating possible training for life work in industrial or 
manual lines according to individual proclivity. 



JAN 2 6 1922- 

Vol. 19, No. 2 

August, 1921 

The Philippine 
Journal of Science 





' ; Published by the Bureau of Scienc^e of the Government of the Philippine Islands 

EtBiBR B. Jteamx, M.S., Editor 

U. C UcGsEGO% A,B., As80ciat€ Editor 

Albert H. Wells, A3.; Giunville A. PeejkinS, PH.D;; A, P. West, Ph.D. 
T- Dak Juan, A.B., Phar.D.; F. Agcaoili, A.B.; A. S. Argublles, B.S. 

Albert E. W. King 

Chemistry ' 

Warren D. Sbhth, Ph.D. ; Roy E. Dickerson, Ph.D. 

H. W. Wade, M.D.; Otto Sch5bl, M.D. 

F. G. Haughwout; Stanton Yotjngberg, D.V.M. 

Experimental Medicine, 

LiSORio Gomez, M.D., Ph.D.; F. Calderqn, B^, L.M. 

Vicente de Jesus, M.D. 

Climcal Medioino 

W. H. Brown, Pfi.D.; C. F. Baker, M.A.; Athbrton Leb, M.S. 

O* A. RBiNiaNG, B.S.A., M.S.; L. M. Gui»RERO, PHAR.D. 


Albert C. Herr^ Ph.D.; C. F. Baker, M.A.; S. F. Light, M.A. 

C. S. Banks, M.A. ; L. D. Wharton, M Jl.; W. Schultzb 


H. 0. Beyer, M.A.; Otto Johns Schebrer, M.A. 

Anna B. Banyea, Copy Editor 

JJwtJstyipt intended tot publication and books tot review should be sent 
to the editor. One hundred s^arates of papers publislied in the Journal 
isre f uniished to authors without charge. Additional copies may be had at 
H^ors' i^cpense. : 

Publications sent in exchange for the Philippine Journal of Science 
should be addressed: Library, Bureau of Science, Manila, P. h 

The Journal is issued twelve times a year/ The subscription price is 6 
dollars. United States currency, per year. Single numbers, 60 cents each. 

Stdmcriptions may be sent to the BusiifBSS Manager^ Philippine Jour- 
nal of Science, Bureau of Science, Manila, P. I., or to any of the agents 
UstM on the third page of this cover. 

The Philippine 
Journal of Science 

Vol. 19 AUGUST, 1921 No. 2 


, By H. Atherton Lee^ 

Mycologist, Bureau of Science, Manila^ 



Field investigations on citrus canker in the continental United 
States have been impossible in the past because of the regula- 
tions which have been promulgated to make eradication work 
possible. For this reason, investigations on citrus-canker con- 
trol were undertaken in the Philippine Islands in the summer 
of 1917. The primary purpose of these experiments was to 
determine whether citrus canker could be controlled by means 
other than total eradication; that is, by the use of fungicides 
and cultural methods. It was believed that, even though 
such control methods should not prove successful, methods and 
ideas might arise which would be of value in aiding citrus-canker 
eradication in the southern United States. 

^ The writer wishes to express appreciation to Col. Adriano Hernandez, 
director, and Mr. S. Apostol, chief of the plant industry division of the 
Philippine Bureau of Agriculture, for the great assistance and the many 
facilities afforded him. It is to them that thanks are due for the use of 
the citrus collection at Lamao, Philippine Islands. Similar deep appre- 
ciation is expressed to Prof. C. F. Baker, dean of the College of Agriculture, 
University of. the Philippines, for the use of the citrus collection at Los 
Baiios during these experiments and for assistance in many other ways. 
Thanks are also due to Father M. Selga, of the Weather Bureau, for very 
kindly assistance in supplying climatological data. 

^ The work was begun while the writer was under the direction of the 
United States Department of Agriculture, and credit is due to that insti- 
tution for the inauguration of this work. 

180732 129 

130 The Philippine Journal of Science 1921 

The writer wishes this paper to be considered as merely a 
progress report, since past work has been of value mainly in 
indicating the great differences in susceptibility of the commer- 
cial Citrus species and varieties and the differences in effort 
necessary for the control of canker on such different species 
and varieties. Such methods of control as have been success- 
ful to some extent are also reported here. 


The citrus industry in the Philippines consists for the most 
part in the production of mandarin oranges (Citrus nobilis) from 
seedling trees; a few seedling sweet-orange trees (Citrus 
sinensis) , a few pummelo trees (Citrus maxima) , and calamondin 
(Citrus mitis) trees are also grown. The largest center of 
production of citrus fruits is Batangas Province, south of Manila. 

In some of the outlying islands citrus canker has not been 
found) as yet, which has given rise to the suggestion of Mackie 
(14) that canker has not existed in the Philippines for a great 
period of time. Although Wester (20) has shown the presence 
of citrus canker in the isolated Batanes (the small islands lying 
between Luzon and Formosa) as early as 1909, he has found it 
to be entirely absent from certain regions of Mindanao and 
Sulu Archipelago. This, then, would corroborate the sugges- 
tion of Mackie. For the purpose of this paper, in any case, 
it is sufficient to note that canker is distributed throughout 
Batangas and Laguna Provinces and the citrus plantings of 
the Manila Bay region where the experiments were carried on. 

The main control work was begun at two places; in the 
orchards of the Lamao agricultural experiment station of the 
Philippine Bureau of Agriculture, and in the citrus plantings, 
in Los Banos, of the College of Agriculture of the University 
of the Philippines. Control work was not attempted on the 
plantings of private holders since they consisted for the most 
part of mandarin orange trees which are in most cases resistant 
to citrus canker. 


An understanding of the climatic conditions at Lamao and 
Los Banos is essential before undertaking a discussion of the 
control work. 

Briefly, at Lamao and Los Banos, the year is divided into two 
seasons ; the wet season, beginning generally about the middle of 
May and ending in November, and the dry season, beginning the 
latter part of November or early December and lasting until the 


Lee: Citrus-canker Control 


early part of May. Table 1, compiled from the reports of the 
Philippine Weather Bureau, shows the monthly rainfall in detail 
during 1914, 1915, and 1916 at Manila. The rainfall at Lamao 
is, sometimes at least, slightly less than that at Manila, while 
that at Los Bafios may be said to be usually slightly greater. 

Table 1. — Rainfall recorded by the Central Observatory at Manila of the 
Philippine Weather Bureau, 


January. __ 
February . 






August --_ 
October ___ 
December - 











































































A study of this table will give an idea of the time and amount 
of rainfall throughout a normal year. 

Table 2. — Temperatures recorded by the Central Observatory at Manila 
of the Philippine Weather Bureau, 



February-- - 


April -- 





September --. 





































































































































The temperatures at either Lamao or Los Baiios are fairly 
high, with comparatively slight seasonal variations. The table 


The Philippine Journal of Science 


of temperatures (Table 2) is also taken from the reports of the 
Weather Bureau of the Philippine Government, and represents 
Manila temperatures. 

The temperatures at Manila may be taken as an index of the 
temperatures at both Los Banos and Lamao. The temperatures 
at Lamao and at Los Baiios, therefore, are probably favorable at 
all times for the dissemination and the development of citrus 

The humidity remains fairly high throughout the months of 
the dry season as is shown by Table 3. 

Table 3. — Relative humidity at Manila, compiled from the annual reports 
of the Philippine Weather Bureau, 
















Per ct. 


Per ct. 

Per ct. 


Per ct. 

Per ct. 



Per ct. 

Per ct. 


Per ct. 

Per ct. 

March __. 



June _ 


August . _ 



November __ 


Although the humidity and the temperature are favorable 
throughout the dry months, it is apparent that canker develop- 
ment is inactive and passive during these months. The explana- 
tion for this inactivity, corroborated by close field observation, 
is that citrus canker apparently is dependent for its dissemina- 
tion upon free moisture on the leaves, in the form of either 
rain or dew. The dry season from December to May, there- 
fore, is a limiting factor in the development of citrus canker in 
the Philippines. The long days of steady, direct sunlight during 
the dry season are also apparently a limiting factor to the spread 
of the canker organism. 

The dry season is correlated more or less with a steady north- 
east monsoon, while the wet season is correlated with a south- 
west monsoon. The velocity of these winds can be judged from 
Table 4, also compiled from the reports of the Philippine 
Weather Bureau. 


Lee: Citrus-canker Control 


Table 4. — Representative wind velocities in Manila, expressed in kiloTneters 
per hour, compiled from the annual reports of the Philippine Weather 
































































April. .__ ._.. 


June __ 






December _ 

During the summer season cyclonic disturbances are frequent, 
when wind velocities of 60 or more kilometers an hour are not 
unusual, and even a velocity of 195 kilometers an hour has been 
recorded by the Weather Bureau. Such winds are usually ac- 
companied by rainfall and are a serious handicap in combating 
citrus canker, in as much as they disseminate the canker or- 
ganism at a time when all conditions favor the development 
of canker. More extensive data and references upon the climate 
of the Philippines may be obtained in the publication of Father 
Coronas, (4) of the Philippine Weather Bureau. 


The available literature on investigations upon the control of 
citrus canker is not extensive. The first note of the effect of 
spraying against citrus canker is that of Wester, (18) who, in 
the dry season of 1912 in the Philippines, was able, by applica- 
tion of Bordeaux mixture, to control a disease which he regarded 
as citrus scab. In the wet season, however, he reports his 
spraying methods to have been unsuccessful against the disease. 
He later identified the disease as citrus canker instead of citrus 
scab. Stevens (16, 17) has reported spraying conducted by grow- 
ers and nurserymen against canker in Florida to be impractic- 
able, but has not described the spraying methods employed. 
Wolf (21) described control methods using variations of Bor- 
deaux mixture against the disease on grapefruit trees in 

134 The Philippine Journal of Science 1921 

Alabama. His conclusions are best obtained by a direct quota- 
tion from his publication : 

All visible signs bf canker were carefully removed from the trees prior 
to the application of the mixtures. Bordeaux mixture, Bordeaux mixture 
and bichlorid of mercury (12 tablets in 3 gallons), Bordeaux mixture and 
formaldehyde (1:100), and a Bordeaux and lead arsenate mixture were 
employed. Applications were made on March 26, April 29, and May 14, 
and no new infections had developed on any of the sprayed or unsprayed 
trees by the last named date. On May 27, however, new infections were 
apparent and were equally numerous on sprayed and check trees. 

These results of course apply only to the grapefruit, the most 
susceptible host. 

Doryland(6) has reported successful results of spraying ex- 
periments against citrus canker upon seedling trees of calamon- 
din, mandarin, and sweet-orange varieties in nursery rows at 
the Singalong experiment station in Manila, Philippine Islands. 
He found that spraying with Bordeaux mixture 4-6--50 met with 
little or no success when used in the rainy season, but later 
he obtained favorable results by using formalin in a 1 to 
100 solution at ten-day intervals. Doryland does not give his 
dates of spraying, but a survey of his statements as to time of 
application leads to the conclusion that his favorable results with 
formalin were obtained in the dry season, when active dissemina- 
tion of citrus canker is limited. His later report (7) states that 
applications at ten-day intervals of Bordeaux mixture 4-6-50, 
plus formalin to make a 0.4 per cent solution, were able to rid 
the plants of canker in four months. His spraying was begun 
January 29 ; therefore, four months would bring his experiments 
to a conclusion on May 29. January, February, March, April, 
and May are dry months in the Manila Bay region, as Table 1 
has shown. By using creolin-formalin emulsion, Doryland was 
able to decrease canker during the first four months, but during 
the fifth month there was a bad increase in canker. The fifth 
month would be June, which has a very heavy rainfall. Thus 
it would seem as if Doryland's data may be taken to show that 
canker can be controlled in the dry months, but that his results 
do not apply to the rainy season when canker diss.emination 
and development are active. This conclusion would bear out 
Westerns result. 

Doidge(5) reports that citrus canker was controlled with 
Bordeaux mixture 4-4-50 under conditions in the Transvaal, 
where there is a long dry season. She does not give the host 
upon which the control was obtained. 

Kellerman(8) advised spraying healthy trees with a 1 per 

19,2 Lee: Citrus-canker Control 135 

cent formalin solution to avoid infection, but the recommen- 
dation apparently was not based upon experimental evidence. 
Spraying experiments upon Washington navels in Japan have 
been reported by several Japanese investigators and were 
apparently successful to some degree as practiced by the inves- 
tigators. Growers of Washington navels in Japan at present, 
however, have given up such spraying as unsuccessful in a 
number of cases. A more detailed review of the Japanese liter- 
ature will be presented in a later publication. 



Lime sulphur solution, Bordeaux 4^-50 mixture, and Bur- 
gundy 3-3f-50 mixture ^ are so commonly used that no discus- 
sion as to the methods of preparation is necessary. During 
the spraying experiments it seemed desirable to attempt to 
render the copper more readily available for action against the 
canker organism, following the theory^ of Bedford and Pick- 
ering; (1, 2) in the case of Bordeaux mixture the excess of lime 
was, therefore, reduced to just the amount sufficient entirely 
to precipitate all of the copper. This was called neutral 
Bordeaux mixture. A similar neutral Burgundy mixture was 
employed in which the sodium carbonate added was just 
sufficient to precipitate the copper with no excess remaining. 
Ammoniacal copper carbonate solution, when used, was made 
up to contain: Copper carbonate, 5 ounces; concentrated am- 
monium hydroxide, t pints; and water, 50 gallons. The 
methods of preparation are, of course, well known. Formalin 
as a preventive was used as a simple solution, easy to pre- 
pare ; neutral lead arsenate is also so commonly used as to need 
no discussion. 

Two different oil-emulsion preparations were used in check- 
ing scale-insect increases following the fungicide applications. 
An oil emulsion described by Yothers(22) was commonly used. 
This consisted of soft soap, 4 pounds; paraffin oil (25° Baume), 
1 gallon; and water, 1 gallon. The preparation was made up 
in the ratio of 1 gallon of the mixture to 50 gallons of water. 
Another oil emulsion consisted of a mixture of a cresylic soap 
liquor with kerosene or distillate. This emulsion was prepared 

' The spraying experiments reported here were begun and carried to a 
conclusion before the work upon the phenol coefficients of fungicides (11) 
with the canker organism had been completed. 

* These references have not been available in Manila but are quoted from 
their reviews. 

136 The Philippine Journal of Science 1921 

as follows: Liquor cresolis compositits, 1 quart; kerosene, 3 
quarts ; and water, to make 50 gallons. 

A sticker was used in many instances in an attempt to cause 
the copper sprays and lime sulphur to adhere to the foliage for 
the greatest length of time possible. This sticker consisted of 
2 pounds of resin, dissolved with heat in a solution of 1 pound 
of sodium carbonate in 1 gallon of water. The sticker was used 
in these experiments in the ratio of 1 quart of the mixture to 
50 gallons of the spray mixture. 

The removal of the sources of canker infection was obtained 
in some cases by pruning out twig and limb cankers. Another, 
somewhat drastic method was used in a few cases of trees 
with heavy foliage infections. A solution of formalin was pre- 
pared which would partially burn the citrus foliage ; leaves with 
canker infections were sometimes already weakened,, and these 
dropped after such a drastic spray, while normal leaves in many 
cases would survive this treatment. A 1 to 80 formalin solution 
was used for this purpose on sweet orange, grapefruit, and 
mandarin orange trees with some degree of success. Such a 
concentration was, however, too strong for lime and lemon trees, 
and almost complete defoliation would follow on those hosts. 
The use of formalin solutions for this procedure was closely 
related to the sunlight, conditions; on rainy days or toward 
nightfall the formalin would not evaporate as rapidly as in direct 
sunlight and greater injury would result. The use of a spray 
solution for this purpose was called a clean-up spray, in contra- 
distinction to the term preventive spray. 

It was also attempted to secure the growth of new foliage 
during climatic seasons unfavorable to the development or dis- 
semination of citrus canker. Such growth was secured by 
employing the pruning procedures at the end of the rainy season 
so that new growth stimulated by the pruning came out in the 
dry season. Maintenance of foliage growth was also secured in 
the dry season by irrigation. 

The test of spray mixtures was made along two lines ; namely, 
in connection with the pruning and the stimulation of growth 
measures, at Lamao, Bataan, Philippine Islands ; and the efficacy 
of spraying alone, without pruning or other measures, was tested 
at Los Bafios, Laguna, Philippine Islands. 


The orchard at Los Baiios is maintained by the College of 
Agriculture of the University of the Philippines and consists of 

19,2 Lee: Citrus-canker Control 137 

rows of native sweet oranges, calamondins, lemons, Kusaie 
limes, citrons, and pummelos. With the exception of the Kusaie 
limes, these trees are largly native to the Philippines, without 
classification into horticultural varieties. The trees were, for 
the most part, five years old at the beginning of the experiment, 
and although varying considerably in size were commonly 3 to 
5 meters in height and, with a few exceptions, in good growing 
condition. The trees were in orchard formation upon level land, 
protected but partially from the strong and sometimes violent 
winds. The orchard was cleian cultivated at all times of the 
year; no irrigation was practiced during the dry season. 

The orchard was divided into six plats, so arranged that each 
plat cut across the rows of similar species at right angles ; each 
species was thus represented in each plat. It may seem unfor- 
tunate that the experiments were not carried on in,' orchards of 
a single uniform susceptible species, having plats consisting of 
but one variety; however, this was impossible in as much as no 
such mature orchard existed at that time in the Philippines. 

The use of the specific name Citrus maxima (C grandis, C, 
decumana) as pointed out by Merrill (15) has been followed in 
this paper. 


The treatment of this plat consisted of an application of lime 
sulphur (32° Baume) in a 1 to 40 solution. It was found that 
lime sulphur washed off very easily during the heavy rains at 
Los Bailos.^ In as much as the rains are normally very heavy 
in the wet season in Los Banos, it was apparent that lime 
sulphur could not be used against canker in this locality and its 
use on this plat was, therefore, abandoned. 


Plat II consisted of two rows running parallel with rows of 
Plats I, III, IV, V, VI, and was left entirely untreated as a 
check upon the rows that were treated. The amounts of canker 

^ In this connection various methods were attempted to cause lime sul- 
phur to adhere to the foliage for a longer period. Powdered casein in the 
ratio of 4 ounces to 50 gallons was sifted into the solution; compared with 
an application of lime sulphur without casein, made at the same time, no 
advantage was gained. Casein in the form of condensed milk was added 
to lime sulphur and similarly gave no advantage. The resin salsoda 
sticker was added in the ratio of 1 quart to 50 gallons of the lime sulphur. 
This mixture was applied to a plat of trees while at the same time lime 
sulphur with no such sticker was applied to a similar plat of trees. The 
foliage of both plats tested for sulphur at the end of two weeks was en- 
tirely negative. 


The Philippine Journal of Science 


on each tree in August, 1917, at the beginning of the experi- 
ments, and in August, 1918, when the experiments were com- 
pleted, are shown in Table 5. 

Table 5. — Showing degree of canker affection of trees in untreated plat, 
Plat II, of the College of Agriculture citrus planting at Los BanosJ^ 


Leaves affected. 

August, 1917. 

August, 1918. 

Citrus maxima . . _ _- . 

Per cent. 



Per cent. 


Do. _ .._ — . 



Do ._ 

Citrus sinensis 



Do. . . , . 





Citrus Tnitis 



- _ - - 

Citrus limonicL 




Do. _ _ 

Do - ._ 

Do. __ _ _ 

Do . 



Citrus hystrix _. ._ .- 



Citrus aurantiJolicL 

Do. _ - 


Do . .. - ... 


Citrus fnedica 












Citrus maxima 






Citrus nobilis 

Citrus maxima. _. _ _. . . 


Citrus maxima . . 







Citrus limonia 




Citrus sinensis 


* The determination of the species mentioned in this and subsequent tables is taken 
from a chart of the college orchard at Los Banos made available through the kindness 
of Mr. Mariano G. Medalla, instructor in the College of Agriculture, University of the 
Philippines. The measurement of the amount of canker on the leaves at Los Baiios in 
August, 1918, is also largely to be credited to Mr. Medalla. For this and for much other 
assistance the writer wishes to express his sincere thanks to Mr. Medalla. 

b Dead. 


Lee: Citrus-canker Control 


It can be seen from this tabulation, that the amount of canker 
in this plat at the end of the experiment in August, 1918, was 
quite as great as it was in August, 1917, when the work was 
undertaken. The criticism may be raised that the estimation of 
the amounts of foliage cankered could not be accurate. This 
is, of course, a valid criticism, but no other method of measuring 
the amounts of canker was available in as much as few or no 
fruits were formed in the orchard during this season. The 
amounts of canker were determined only after long and careful 
examination and, in the case of the affected leaves expressed 
numerically, is from actual count. 


The treatment of this plat was as follows : 
August 24, 1917. Cresol-kerosene emulsion, plus mercuric bichloride 

to make a 1 to 1,500 solution. 
September 14, 1917. Burgundy 3-31-50 mixture. 
October 11, 1917. Burgundy 3-31-50 mixture. 
November 12, 1917. Burgundy 3-31-50 mixture, plus powdered 

neutral lead arsenate to make a 1-50 mixture. 
December 20, 1917. Burgundy 3-3-50 mixture. 
January 5, 1918. Yothers's oil emulsion. 

May 17, 1918. Neutral Burgundy mixture, plus formalin 1-100. 
June 14, 1918. Neutral Burgundy mixture, plus formalin 1-100 and 

cresol-kerosene emulsion. 
July 13, 1918. Neutral Burgundy mixture. 

Table 6. — Showing degree of canker affection of trees in Plat III of the 
College of Agriculture citrus planting at Los Banos. 


Leaves affected. 

August, 1917. 

August, 1918. 

Citrus maxima _.. 

Do — - 

Per cent. 






Per cent. 









Do _. 






Citrus mitis __ . 





Citrus limonia 






The Philippine Journal of Science 


Table 6. — Showing degree of canker affection of trees in Plat III of the 
College of Agriculture citrus planting at Los Banos — Continued. 


Leaves affected. 

August. 1917. 

August, 1918. 

Citrus Zimonta— Continued. 

Do i_ 

Per cent. 


Per cent. 










Do.. ■ .._ 






Do- . 



Citrus hystrix.. .. ______ 



Do - 

Citrus cLurccntifolici 





Do- . 


Do - - . 






Citrus medico. 


Do - 


Do - 




Citrus TncLxima - - 







Citrus medica .. . -- 

Citrus hystrix 


Do . _ 




















Citrus litiionia, 


Do— -._ - 



19,2 Lee: Citrus-canker Control 141 

Cresol-kerosene emulsion plus mercuric bichloride 1 to 1,500 
was used as an insecticide at the beginning of the experiment, 
to check the increase of scale insects. It was found that the ad- 
dition of bichloride to make a 1 to 1,500 solution was much 
too strong a mixture, and defoliation of many of the normal as 
well as cankered leaves resulted. The neutral Burgundy mix- 
ture was employed under greatly varying weather conditions; 
upon the citrus trees no evidence of burning of the foliage or 
fruit was ever observed in these experiments. 

The foregoing tabulation shows that a reduction in the 
amounts of citrus canker was obtained in this plat. The most 
noteworthy cases are those of the Kusaie limes and the trees 
of Citrtcs limonia, both of which species were reduced from a 
condition of fairly general infection to a condition of fairly 
satisfactory control. 


The treatment of this plat was as follows : 

August 24, 1917. Bordeaux 4-4-50 mixture plus sugar (sugar, 15 

per cent by weight of the copper sulphate in the mixture) and 

plus 1 quart of resin, sal-soda sticker. 
September 13, 1917. Bordeaux 4-4-50 mixture plus sugar (sugar, 

15 per cent by weight of the copper sulphate in the mixture) and 

plus 1 quart of resin, sal-soda sticker. 
October 11, 1917. Bordeaux 4-4-50 mixture plus sugar (sugar, 15 

per cent by weight of the copper sulphate in the mixture) and 

plus 1.25 quarts of resin, sal-soda sticker. 
November 12, 1917. ^Bordeaux 4-4-50 mixture plus sugar (sugar, 15 

per cent by weight of the copper sulphate in the mixture) and 

plus 1.25 quarts of resin, sal-soda sticker. To this was also 

added for this application powdered, neutral, lead arsenate to 

make a 1-50 mixture. 
December 20, 1917. Neutral Bordeaux mixture. 
January 3, 1918. Yothers's oil emulsion. 

May 16, 1918. Neutral Bordeaux mixture plus formalin 1-100. 
June 14, 1918. Neutral Bordeaux mixture plus formalin 1-100 and 

cresol-kerosene emulsion. 
July 13, 1918. Neutral Bordeaux mixture. 

The use of sugar in Bordeaux mixture is described by Bour- 
cart(3) as having been first suggested by Perret. It was said 
to make the copper more easily available and to cause the Bor- 
deaux to adhere better to the foliage. Lutman(l3) states that 
the addition of sugar to the lime before preparing Bordeaux 
mixture also weakens the precipitation membranes, causing the 
precipitation to go more nearly to completion and increasing 
the covering power of the mixture. 


The Philippine Journal of Science 

Table 7. — Degree of canker affection of trees in Plat IV of the College 
of Agriculture citrus planting at Los Banos, 


Leaves affected. 

August. 1917. 

August, 1918. 

Citrus maxima 

Per cent. 













Per cent. Number. 

Do .- 






4 i 





n , 

Citrus sinensis . . _ 

Do __ _ 

Do - 



Do — 

Do - - - - 

Citrus TTiitis -_ 





Do - - 

Do - 

Citrus limonia . . 

• 1 


Do — .— 


Do — 


Do — 


Do — 

Do - 


Do - - -- -- 


Citrus hystrix - .. 

Do . - - 


Do -- .-- . - 

Citrus aurantifolia 

Do — 

Do - _ 

f^ 1 






Do --. - - 




Do . 


Citrus medica - -_ . 





Do- - 


Citrus m.axima 

Do - 



Do- — 





U . j 

a Dead. 

19, 2 

Lee: Citrus-canker Control 


Table 7. — Degree of canker affection of trees in Plat IV of the College 
of Agriculture citrus planting at Los Banos — Continued. 


Leaves affected. 

August, 1917. 

August, 1918. 

Citrus hystrix 

Per cent. 


Per cent. 






Do ._.. 

Citrus maxima 

Do -- 








It is apparent that some reduction in the amounts of canker 
infection was accomplished in the foregoing plats. 


The trees in this plat were of the same species and varieties 
as those represented in Plats I, II, III, and IV. This plat was 
left entirely untreated, as a check upon the sprayed plats. 

Table 8. — Degree of canker affection of untreated trees in Plat V, of the 
College of Agriculture citrus planting at Los Banos, 


Leaves affected. 

August, 1917. 

August. 1918. 

Citrus maxima . 

Per cent. 



Per cent. 







Citrus mitis 



Citrus limonia _ 


Do ... 











Do .._. 



Citrus hystrix 

Citrus aurantifolia 









Citrus maxima 

Do .._ 


144 The Philippine Journal of Science 1921 

Table 8 shows that in this plat canker, even though untreated, 
decreased to some extent from what it was at the beginning of 
the experiment. Although there are possible explanations for 
this they will not be entered into in detail here; granted that 
seasonal differences permitted some reduction in canker, it is 
apparent that considerably more reduction was obtained on 
the sprayed plats. 


It was intended to run this plat entirely as a test of formalin 
against citrus canker. It became evident a few weeks after 
the first application, however, that new infections were appear- 
ing and that applications of formalin as a preventive would be 
necessary much too often to be feasible. The spraying with 
formalin solution was therefore abandoned as a preventive but 
was subsequently used on other plats for clean-up work. 


The comparative value of the sprays which were tried is 
briefly summarized as follows: Cankers recurred after the ap- 
plication of formalin 1 to 100 without any reduction in number 
or distribution; the results at Los Baiios are therefore taken 
to indicate that formalin has little value as a preventive spray. 
In the heavy tropical rains in the Philippines lime sulphur did 
not adhere well to the foliage, although it is possible that in 
regions of more moderate rainfall lime sulphur would be of 
great value. Burgundy and Bordeaux mixtures, when freshly 
prepared, adhered to the foliage very well and no difficulty was 
experienced with the washing of such copper sprays from the 
foliage. The determination of the length of the periods between 
spraying depended, therefore, chiefly upon the amount of new 
growth which came out and was unprotected by coatings of spray. 
As to the comparative merits of Burgundy and Bordeaux mix- 
tures as preventives, little or no difference showed in the results. 

From Tables 6 and 7, showing the results of plats sprayed 
with Burgundy and Bordeaux mixtures, it is apparent that the 
amounts of citrus canker in most cases were reduced to some 
extent; Table 5, showing an untreated plat, shows no decrease 
in canker; Table 8, also untreated, destroys some of the value 
of the results on the treated plats, since it also in some cases 
shows a decrease in the amount of canker affection. There' 
are possible explanations for this ; but, regardless of such expla- 
nations, it is apparent that the reduction in the amounts of 
canker in the treated plats is greater than in either of the 
untreated plats. 

19,2 Lee: Citrus-canker Control 145 

It would seem, then, that such preventive sprays have, to some 
extent, reduced the amounts of infection. In the light of the 
writer's disinfectant tests (H) it is difficult to ascribe the cause 
for this reduction unless it be to the excesses of copper precip- 
itants applied, or even possibly to the mechanical effects of 
the spray deposits on the foliage. 

The criticism of the results, from a commercial viewpoint, 
would be of course that nine spray applications are not eco- 
nomically feasible. It should be noted however that, theoret- 
ically, control should more nearly approach the absolute, year 
after year, since with the gradual prevention of canker infec- 
tion by preventive coatings there is also a gradual reduction 
in the sources of infection. This would reduce to some extent 
the necessity for such frequent spray applications. It would 
seem probable also that, with greater knowledge of the climatic 
peculiarities of the locality and the growth periods of the dif- 
ferent varieties, the number of spray applications would be re- 
duced somewhat. This can only be determined by a continuation 
of the experiments, which will be reported upon in further 



In these experiments no effort was made to compare the value 
of the different spray mixtures used, the main purpose being 
to determine whether or not canker control was feasible by any 
and all means. The experiments were carried on in three or- 
chards of the Lamao experiment station of the Philippine Bureau 
of Agriculture. These orchards consist of collections of the 
common horticultural varieties of Citrus species grown in the 
United States and Japan, usually two individuals representing 
each variety.^ 

The division of these orchards into spray plats in such a way 
as to include individuals of each species in each plat was difficult ; 
however, an attempt was made to make each plat representative. 

" These collections are mainly the work of Mr. P. J. Wester, agricul- 
tural advisor, Philippine Bureau of Agriculture. The collections are 
very extensive and contain a great number of the commonly grown citrus 
varieties of America as well as many Japanese-grown varieties. In ad- 
dition, Mr. Wester has introductions from India, Siam, China, and Aus- 
tralia, as well as a very extensive collection of the native Philippine 
fruits. The writer is greatly indebted to Mr. Wester for his very hearty 
cooperation in the identification of obscure varieties, the use of his card 
indexes, and all matter relating to his collections. 

180732 2 

146 The Philippine Journal of Science 1921 

It is unfortunate that plats of uniformly susceptible varieties 
could not be employed ; but, as stated previously, there were no 
such mature orchards in the Philippines. 

The trees, except in a few instances, were in good growing 
condition, planted in orchard formation on level, clean-cultivated 
land. The trees were for the most part five years old at the 
beginning of the experiment. The orchards were surrounded 
by dense thickets of bamboo, 10 to 14 meters high, and these 
bamboo thickets formed very efficient windbreaks. 

All plats at Lamao underwent entirely the same treatment 
in regard to irrigation, fertilization, and cultural operations. 
Only treated plats underwent pruning, and no pruning of the 
control plats was made. 

All spraying ijientioned here was carried on with a hand-power 
pump equipped with a pressure tank, and all spray applications 
were made with the pressure maintained at from 120 to 140 
pounds. Operations for the removal of infected parts and the 
stimulation of new growth were identical in all the treated plats. 

The detailed reports will not be presented from all of the 
plats, in as much as the general conclusions from each plat 
were much the same. Several of the plats, however, contained 
varieties, the susceptibility and resistance of which may be of 
interest in America; these plats have been selected for presen- 
tation not only to show the effect of the canker-control methods 
but also to show comparative susceptibility. 


September 3, 1917. Lime sulphur (32° Baume) 1-40 solution. 

September 20, 1917. Bordeaux 4-4-50 mixture. 

September 21, 1917. Trees pruned for twig cankers. 

October 23, 1917. Bordeaux 4-4-50 mixture. 

November 20, 1917. Formalin 1-80 solution. 

November 21, 1917. Trees pruned for twig cankers. 

November 28, 1917. Lime sulphur 1-50 solution plus formalin 1-100, 
and pmvdered, neutral lead arsenate to make 1-50 mixture. 

December 7, 1917. Kerosene-lysol emulsion, plus formalin 1-70. 

January 10, 1918. Neutral Bordeaux mixture plus resin, sal-soda 
sticker and formalin 1-100. Trees carefully pruned for twig 

May 31, 1918. Lime sulphur 1-40 solution. 

June 25, 1918. Neutral Bordeaux mixture plus kerosene-lysol emul- 
sion, plus formalin 1-100. 

August 1, 1918. Lime sulphur 1-35 solution. 

The condition of the trees in regard to canker may be followed 
through the different seasons in Table 9. 

19.2 Lee: Citms-canker Control ^47 

Work on this plat was begun by spraying with lime sulphur, 
but it became apparent, as at Los Banos, that the lime-sulphur 
solution was easily washed from the leaves by the beating, 
driving rains. Thereafter a change was made to Bordeaux 
mixture, except in seasons when the rainfall was less intense. 
In such seasons of less rainfall, because of the increase of scale 
insects following the successive use of copper sprays, lime sul- 
phur was used whenever possible for its action as a scalecide 
as well as for its action against canker. 


This plat consisted of three rows of trees adjacent to the rows 
of Plat I ; the character of the trees and degree of canker infec- 
tion throughout the season are shown in Table 10. All trees in 
this plat were left entirely untreated. Unfortunately these trees 
are for the most part varieties different from those that appear 
in the sprayed plats; nevertheless, they may afford some basis 
for comparison between the treated and the untreated plats. 

Although some of the trees in this untreated plat show re- 
ductions in the amount of canker infection, some show increases. 
As a whole, then, this plat, serving as a control, can be used 
as a basis for comparison with the treated plats. The trees 
being untreated also afford something of an idea of the sus- 
ceptibility of the species represented in this plat. 


This plat consisted of three rows of trees, the character of 
which is shown in the tabulation of the degree of canker in- 
fection during the various stages of control experiments. The 
treatment of the plat was as follows: 

September 1, 1917. Lime sulphur 1-40 solution. 

September 20, 1917. 4-4-50 Bordeaux mixture. Trees pruned for 

twig cankers. 
October 23, 1917. 4-4-50 Bordeaux mixture. 
November 20, 1917. Formalin 1-80 solution as a clean-up spray. 
November 28, 1917. Formalin 1-100 solution plus powdered, neutral, 

lead arsenate to make 1-50 mixture. 
December 6, 1917. Lime sulphur 1-40 solution plus formalin 1-80. 
January 9, 1918. Neutral Bordeaux mixture plus formalin 1-100 and 

Yothers's oil emulsion. Trees pruned for twig cankers. 
May 31, 1918. Lime sulphur 1-30 solution. 

June 24, 1918. Lime sulphur 1-35 solution plus formalin 1-100. 
August 1, 1918. Lime sulphur 1-35 solution plus resin, sal-soda sticker. 

The complete reversal from Bordeaux mixture to lime-sulphur 
solution in 1918 was necessitated by severe attacks of scale in- 
sects following the Bordeaux applications. 


The Philippine Journal of Science 










o o o o 

o o o o c 









o o o c 










to rH 










o o o 

o o o o 

OJ U5 O 











1-t (M tH 













o o o o o o o 

O O O O (N 

OI o o 


g ^ i 



«o o o o o o o 

O O t> rH 

o o 

CO o 















"=' S S "= g g g 1 









o o 


^ ;:: i 






o o o o o o 

^ ° g g 8 

g ^ o g 

§ ^ S g 8 g 



tH rH 


























^ 2 S^ 

?^ § Q 




•*i g 




o ^ o ^ 






c3 S Cj 


































g Q 


Native oi 
Messina . 
Murrill o 





Rough le 




19, 2 

Lee: Citrus-canker Control 


j 1 JO jOOOO j j*^ 


1 1 1 1 j rH O O t- jo 

! 1 ! ! ! 3 ' 

LO U5 lO r-l <N 1 O I I O O 

I 1 1 I 1 iH ! 1 I O 1 O 

ssss"^ is'^'^'^p i 

j j j j I j j j 1 O 1 I 



i i i ; -i 

I i I ! i 

■i I 

s e 

& i i 

O M .2 

^ ^ C 

^ s 

tj. CO W 

o <y >< 

:c "S -9i 

O .2 (y 
to 'tJ 

^ S3 
o '-; M 

S . o 

1^ w 


d c 

^ ^ 


•^ ^ 



T ^ 


?n ^ 

^ 5« 

cu ^ -^ 

O o c 


The Philippine Journal of Science 










1 M M 

t 1 


K i 

«2 i 




C \ \ \ \ 

^ : ; ! : 


1 loo 
1 1 s^ 

1 to 

J w ; 

1 i i 



i M ; 1 

1 : i i i 

1 rH 1 

[ (M 

^LC5U50000U5(N1tH lO'-l rH 1 


-=> 1 1 I 1 '. I 1 

Si!!!!! : 

1 i i i i ! i ; 

1 1 1 

1 CD 

u 11 ! 

1 i i i 





!>■' ill! 
<» 1 1 1 1 

1 M M 

! «o loo 

i i i =5 <=■ '^ 

1 ^ 1 




Citrus hybrid tangelo 


Citrus maxima 


i i i « i -^ 

1 1 1 g 1 cr 
1 1 1 8= 1 & 

. 1 1 > r 

: 1 1 s 1 5e 
1 1 1 ^ 1 '« 

2 ?; 
na T3 -ri g 'O g 


T3 T3 T3 T 


Citrus nobilis 



i J 1 

g Q 1 Q 1 
w !^ H 


Pink pummelo 


Lemon variety 




Washington navel 



i ; ; 
I : 1 


19,2 Lee: Citrus-canker Control 151 

Table 11 shows that in all cases of susceptible trees, a consider- 
able reduction in the amounts of citrus canker took place during 
the year. 

The behavior of the trees of the Tahiti lime is noteworthy. 
Both of these trees were in actively growing condition and well 
exposed to infection; however, they exhibited a very moderate 
susceptibility to canker, and control was very easily obtained. 
Other trees of the Tahiti lime have been under observation at 
Lamao and have exhibited the same slight susceptibility to 
canker. This is all the more peculiar, as Wester (19) also has 
pointed out, since for the most part lime varieties are extremely 

The behavior of the Triumph grapefruit is also noteworthy ; 
as grown at Lamao, it apparently has a less degree of susceptibil- 
ity to canker than have the other American-grown grapefruit 
varieties. The lemon varieties as shown in the table are but 
moderately susceptible and they responded readily to the control 
methods employed. 


The treatment and results were much the same on this plat 
as on plats I and III. 

Of interest from this plat was the ease of control of the Limon 
Real, a Philippine, lemonlike fruit. Of greater interest in the 
United States is the behavior of the Mediterranean sweet orange 
varieties, the Jaffa, St. Michael, and Ruby. The susceptibility 
of these varieties is very slight, and they responded very 
quickly to the control methods employed. 


It was intended that this plat should be sprayed continuously 
with ammoniacal copper carbonate solution as a test of that 
spray as compared with other mixtures. 

At the beginning of October, after two applications, it was 
apparent that this spray was having little effect upon citrus 
canker, in as much as new infections were appearing continuous- 
ly. Spraying was abandoned, therefore, and the plat became a 
check upon the other plats in Orchard B, for the next spraying 

Table 12 shows that on a few trees in this plat the amounts 
of canker decreased; on the other hand, there was an equal 
number of cases where the amounts of canker increased. Ex- 
amination of the table enables one to compare the canker condi- 


The Philippine Journal of Science 













k O O O O O CO 

O O I o o 

o o o o o o o 

«0 O O t- Tf O CO 


o o o o o 

O O (M O rH O 

O O O '-I 

1 loio looooooo 

i?D ! 1 [COOOtOOON 

5.. o o o OS cq o 

1-H O O CVJ tH O O 

2 I .§ ! ! 1 -2 ; ! I i 

« 1 g I I I W I 1 j J5 

S ' O 1 I I I 1 I I "« 

g ! S ! ! ! .g 1 ! 1 e 

M 1 i^ I I ! M ! I ! s 

S rc5 ^ "XJ TS 'O 











Lee: Citriis-canker Control 



oooo loooooooo 

i j 


lo 1 I loooo j loo 

io(N 1 1 joooooiouao 

1 IrHr-li-IOOOO 1 I lo 

I 1 I 1 InOOO iiHOO 

k«U3000 loOOiH lOO 

I I j 1 j j loONOOO 

gssss-^g^^ j i^° 





.2 o 

13 "^ 

Q 2 .^ P 

o i G i S 

£H « C O "_.<-> 

< .2 


The Philippine Journal of Science 







O CO o o 

o o o eg 

<=> ^ S S o o 

o o o o o 

o o o o 




o o o o 




o o o o 

o o o o o 

O kO o o 

O O O W3 O 


\a CO 00 

^ \a to CO la 

J> o o o o o o 

CO O CO 00 OJ o 




i ^ 



i ^ 



1 ■« 



1 1 

s G 

o o o o o 

2 o « 5 o 

c3 S 

ft Q « 


o ^ o 

o o ^ o 5 o 



o « 

o w o 


Q I Q I I Q 




ft o 

% P 

bo Q 


Lee: Citrus-canker Control 


ooo»Hci < leoooo 

o o o cj eo 

(S r-l o o 

O I 00 I to 

o o o o 

O O O W 

t-MOO O I lO > i£> iM 

O O O f-f 

O O O O .-I o o 

O I p o 

oooouaoioia looiooooor-iooo 

e g a 

u o o 

o o 

60 «0 

1 1 

i3 S 

.^ Q 


.t Q 

S § S H 

Q !z; 

Q !: 

M Oh 

156 ^^^ Philippine Journal of Science 1921 

tion of the untreated trees with the amounts of caliker of the 
other, treated plats. 

The trees of this control plat also afford a fair idea of the 
susceptibility of the horticultural varieties represented. The 
Parson Brown, Homosassa, and Pineapple, sweet orange varieties 
of Florida origin, show a considerably greater susceptibility than 
do the Valencia, Majorca, Maltese, and Du Roi, which are 
Mediterranean varieties. The King, Suntara, and Szinkom, 
varieties of the mandarin orange, illustrate the almost entire 
freedom from canker of that class of hosts. The great 
susceptibility of the American-grown grapefruit varieties is 
illustrated by the trees of the McCarty variety in this plat. 


The rows of trees of this plat paralleled and were contiguous 
to those in Plat VI ; treatment was as follows : 

September 4, 1917. Cresol 1-80 solution. 

September 20, 1917. Trees pruned for citrus canker. 

September 22, 1917. Lime sulphur 1-40 solution plus formalin 

October 22, 1917. 3-30-5O Burgundy mixture plus resin, sal-soda 

November 21, 1917. Formalin 1-80 solution plus powdered, neutral, 

lead arsenate to make a 1-50 mixture. 
December 1, 1917. Formalin 1-80 solution plus powdered, neutral, 

lead arsenate to make a 1-50 mixture. 
December 4, 1917. Trees pruned for twig cankers. 
December 7, 1917. Lime sulphur 1-40 solution plus formalin 1-100. 
January 10, 1918. 3-3-50 Burgundy mixture plus Yothers's oil 

January 17, 1918. Trees pruned for twig cankers. 
June 1, 1918. Lime sulphur (25° Baume) 1-30 solution. 
June 26, 1918. 3-3-50 Burgundy mixture plus cresol-kerosene emul- 
sion and formalin 1-100. 
July 30, 1918. 3-3-50 Burgundy mixture plus cresol-kerosene 


Cresol solution was used at the beginning of the work as a 
clean-up spray; however, it had no value in this respect when 
used at a dilution of 1-80 and its use was abandoned. In the 
succeeding spray applications, lime-sulphur solution was used at 
different times to check the increase of scale insects, but the plat 
was in the main sprayed with Burgundy mixture. 

Table 13 indicates that citrus canker has been materially re- 
duced in amount upon all affected varieties and species. As 
evidenced in the table, the difficulty of control of the Duncan, 
the Marsh, and the McCarty grapefruits and the Everglade and 

19,2 Lee: Citrus-canker Control 157 

the Trinidad limes is noteworthy. This can be compared with 
the relative freedom from canker and ease and quickness of 
control of the Mediterranean sweet orange varieties, the Du Roi 
and Hart's Late. 

Plat VIII, in Orchard B, yielded much the same conclusion and 
need not be presented in detail here. In this plat the Paper 
Rind and the Maltese, varieties of the sweet orange, showed 
slight susceptibility and responded readily to the control 
measures. These varieties are of interest since they are usually 
also classed with the Mediterranean group of sweet-orange 


From the foregoing it is apparent that canker has been re- 
duced in the treated plats to a degree at which it may be said 
that a reasonable control has been obtained. Untreated plats 
have shown no reduction in canker. Of course, the number of 
spray applications must be considered, ten fungicide applications 
having been made on those plats and a considerable expenditure 
of labor incurred in pruning. Although such procedure would 
immediately appear prohibitive as an orchard practice, to arrive 
at the correct conclusion as to its feasibility consideration must 
be taken of the different citrus species as hosts, of the different 
degrees of their susceptibility and the bearing such susceptibility 
has upon the ease or difficulty of control, and of the cost of con- 
trol. The spray applications and other methods used on these 
plats sufficed to decrease the amounts of canker, to a degree that 
could be considered a control, upon the most susceptible hosts, 
the lime and grapefruit varieties, as well as upon the i^ore 
moderately susceptible hosts. Upon less susceptible hosts, such 
as the sweet orange, the lemon, and the mandarin orange, these 
control measures apparently could be much simplified and, 
consequently, of course be made cheaper. 

The correct conclusion from these experiments would seem to 
be that control has been obtained, but that upon the grapefruit 
and limes such control is not economically feasible by the methods 
employed; upon the less susceptible hosts control has also been 
obtained, though the treatments applied apparently exceeded 
actual requirements for the efficient control of canker. There- 
fore, no conclusions as to the practicability of control upon such 
hosts can be drawn as yet. Further experiments are now in 
progress to make possible definite conclusions on the control of 
such moderately susceptible hosts. 


The Philippine Journal of Science 












o o o o o o 

o o t>- o o o 


O O CO o o 


I § S S g g 


O rH O O O O 
tH 1-1 CO 

o o o o 

N O O U3 


O O »0 U5 

o o o •^ o o 

OS o o CO 


.s I .g 

i ! I 

1 ii 

O ! O O O 

O !0 !0 !0 !0 !0 !0 !6 ICJ 


P euSw W c« ;z; ;z; 




Lee: Citrus-canker Control 




U5 O O 1-1 O O O 

lo I looooio loo looo 

u. CO ^ 1 cs, o o 1 

O O O tH O O o 

1 ! ikO loOOlO lOO jOOO 

I ! 1 j j I o o 

1 o o I 1 1 I 

O O 1 I o o 









73 ^ro p'O r'O pTS r'O f^TS 

^ is i cj i ^ 10 : c 

« o 
> Q 


Q .£ Q g (S a a -3 , 


g Q 

160 The Philippine Journal of Science 1921 

The observations presented in the tables were supplemented by 
a less complete and thorough observation, made in October, v^hen 
it v^as found that, although some slight increase in canker had 
taken place in the case of the grapefruit varieties, the varieties 
of sweet orange and less susceptible species had maintained their 
comparative freedom from canker. The control obtained, there- 
fore, cannot be criticized as having been secured during seasons 
unfavorable to the activities of the citrus-canker organism, 
although advantage was taken of such an unfavorable period 
(the dry season, December to May), to minimize the dangers of 
reinfection during favorable periods (the wet season, June to 
November) . Such control as was obtained, in other words, was 
secured in the dry season and, except in the case of some of the 
grapefruit varieties, was maintained through the wet season, 
from June to October. 

Preventive sprays. — As to the comparative value of the dif- 
ferent spray mixtures as preventives, little can be concluded 
from the work at Lamao, because of the varied program for each 
plat. However, certain other factors developed which would 
eliminate or qualify the value of some of the mixtures. Thus 
copper sprays, if unaccompanied by scalecides, caused a great 
increase in scale insects ; lime sulphur probably could be used in 
regions of less intense rains, and in such cases would be of the 
greatest value because of its additional function as a scalecide ; 
formalin 1 to 100 apparently was of little or no value as a 
preventive. « 

The foregoing being the conclusions derived from the experi- 
ment, it also seems desirable to offer several suggestions for 
control procedures that became evident and that may be of value 
in future work. 

Clean-up sprays. — In many cases of heavily infected trees, a 
spray of formalin 1 to 80 caused a large proportion of the pre- 
viously weakened cankered leaves to drop ;' in other words, for- 
malin 1 to 80 to some extent seemed to be successful as a selective, 
clean-up spray. On trees already well infected such a clean-up 
spray, followed by careful pruning out of canker, materially 
hastened control. 

Stimulation of growth at climatic periods unfavorable for 
canker development. — Another more important point was the 
stimulation of new grovrth at periods in the year when the lack 
of moisture limited the activity of the canker organism. At 
Lamao such new grovrth was stimulated in the early part of 

19,2 Lee: Citrus-canker Control \Q\ 

the dry season, with the result that such foliage matured and 
hardened during the dry season without incurring danger from 
reinfection. This matured foliage then entered the rainy season 
with the danger of infection very much minimized, and only in 
the case of the grapefruit and lime varieties was difficulty 
encountered in the infection of such foliage. 

Control of insects to prevent dissemination of canker, — Along 
the edges of leaves where chewing insects have bitten, cankers 
are very often found (Plate 1). Cankers also often appear in 
definite trails following the work of leaf miners (Plate 2) . This 
emphasized the need for thorough control of such insects, and the 
addition of lead arsenate to the spray mixtures was frequently 
made. In canker-control work on all varieties of Citrus species 
at Lamao, then, it has been evident that insect control is also an 
essential factor. 

Windbreaks to aid in prevention of canker dissendTvation. — 
From field evidence it was apparent that another agent which 
contributes to the dissemination of citrus canker is the wind. 
This aid to canker development is due not alonef to the spread 
of the canker organism by the wind, although that is of course a 
considerable factor, but also to the fact that a strong wind, whip- 
ping the twigs and foliage, causes spine wounds and surface in- 
juries to the leaves resulting in infection by the canker organisms 
under conditions in which they would not develop on unwounded 
tissue. It has been shown in unpublished experiments on the 
moderately susceptible species that canker will develop at a 
wound on a mature or nearly mature leaf, while infection will 
not take place, or at least will not develop, on uninjured surfaces 
of the same leaf. The prevention of the whipping and stabbing 
of the branches and foliage by the wind is therefore also an 
essential point. The windbreaks at Lamao consisted of dense 
thickets of bamboo, 10 to 15 meters high and almost im- 
penetrable. Such windbreaks were very efficient and very 
materially aided citrus-canker control work. 

At Lamao it has been observed that, by attention to such ap- 
parently minor points as the control of chewing insects, the 
utilization of windbreaks, and the regulation of the growing 
periods, much can be done to minimize the development of canker 
on the moderately susceptible hosts. The comparison of results 
at Los Bafios with results at Lamao supports this conclusion 
and would point to these indirect methods of control as being 
fully as important, if not more so, as prevention of the disease 
by spraying. 

180732 3 

162 The Philippine Journal of Science 1921 


It became apparent at Lamao that there are wide differences 
in the susceptibility to canker (9) of the different species of 
Citrus which necessitate a separate consideration for each 
species when working with, or discussing, such subjects as the 
control of canker, the injuries resulting from canker, and the 
eradication of canker. Many of the contradictory statements 
with regard to various phases of citrus canker would seem to 
be due to a confusion of the species or, in some cases, to the 
failure to connect such statements with the host under observa- 
tion. A consideration of the susceptibility of the different 
species at Lamao in connection with the feasibility of control 
methods has been possible and therefore is briefly presented 

At the beginning of these experiments the trees of the 
mandarin orange varieties had little or no canker. Very few 
infections were found, and it was a carefully recorded observa- 
tion that more than 33 per cent of the few infections that were 
found upon mandarin orange varieties occurred at very evident 
wounds, either spine wounds or insect injuries. Under such 
conditions, as shown in the foregoing tables, complete elimina- 
tion of citrus canker upon the mandarin orange varieties was 
very prompt and there were no recurrences of infection; also, 
the operations were simple and inexpensive. Many of the 
citrons (C. medica) were also slightly susceptible and very easily 
freed from canker. The mature calamondin trees were never 
observed in the writer's experience to be cankered, and control 
measures for citrus canker on this host were entirely unneces- 
sary. These species, the mandarin oranges (C. nobilis var. 
deliciosa) , the citron (C medica) together with the calamondins 
(C mitis) and, of course, the round kumquats (Fortunella 
japonica) , constitute a class of commercially grown citrus fruits 
upon which control is simple and very often unnecessary, at least 
when trees are mature. These varieties might be classed as a 
group, class 1, the members of which are so slightly susceptible 
that they can mature and produce fruits with no injury from 
citrus canker, even in the absence of control treatments. 

At Lamao almost complete elimination of canker was obtained 
very quickly on the American-grown lemon varieties (C. limonia) 
and the so-called Mediterranean varieties of the sweet oranges 
(C sinensis) y such as the Valencia, Jaffa, Mediterranean Sweet, 
Ruby, St. Michael, and Du.Roi. In this class should also be 

19,2 Lqq: Citrus-canker Control 163 

placed the Tahiti lime (C. aurantifolia) . In as much as all the 
other limes under observation were found to be highly suscep- 
tible, the relative freedom from canker of the Tahiti lime is very 
striking and should be taken advantage of in commercial growing. 
The species and varieties enumerated might be considered as 
class 2, all of which showed such a quick elimination of canker 
at the beginning of the control attempts that such measures 
would seem to be economically feasible. The Unshiu varieties 
known in America as Satsuma oranges (C. nobilis var. unshiu), 
at least the Ikiriki, Owari, Zairai, Ikeda, and Wase, although not 
shown in these Lamao plats, from later experiments would also 
seem, to belong in this class of susceptible but easily controlled 

A third class would contain many of the varieties of the sweet 
orange originating in Florida, such as the Pineapple, Homosassa, 
Magnum Bonum, Whittaker, and Parson Brown. This class of 
sweet oranges is more susceptible than the varieties of the Medi- 
terranean group and would be more difficult to control. In this 
class would also be placed the Natsumikan of Japan and many 
of the pummelos, or East Indian type of C. rmiocima (grandis). 
Several of the strains of navel oranges would probably also be 
put in this class. The strains of the navel orange exhibit a 
considerable range of susceptibility; that is, some navel orange 
trees have been observed which were but very slightly affected, 
although exposed to infection for a number of years, while other 
strains, after exposure to infection for only one month, showed 
large percentages of affected fruits and foliage. The determina- 
tion of the susceptibility of these various strains of navel oranges 
is an important problem to be solved. To this class may be also 
added the Triumph grapefruit. The Triumph is the only variety 
that has been under observation which has shown any modifica- 
tion from the extreme susceptibility of the American-grown 
grapefruit varieties. It is by no means resistant, but its lesser 
degree of susceptibility to canker would perhaps make it of more 
value for commercial growing, and it might possibly serve as a 
basis for breeding toward resistance. 

Upon the varieties given above in what may be called the third 
class, it has been shown at Lamao that canker can be controlled. 
It is a question, however, as to whether this control is economic- 
ally feasible. The control obtained on these species and varieties, 
as has been shown, was obtained only after very careful pruning 
and a very large number of spray applications. The experi- 
ments have been continued in Japan and it is hoped that a report 

164 ^/^^ Philippine Journal of Science 1921 

can be made on the economic feasibility of such control in one 
or two years. 

The fourth class would include the extremely susceptible 
grapefruit and lime varieties. Good control has been obtained 
on a few of these varieties, while on others the amounts of canker 
have at least been reduced. Such reductions however have been 
secured only after the application of an uneconomical number of 
sprays and much time in the careful pruning out of twig cankers. 
Under Philippine conditions, it is easily apparent that control 
methods on the American-grown grapefruit varieties and on the 
West Indian lime varieties is not economically practical by the 
methods employed. With this class of citrus varieties therefore 
the results are similar to those obtained by Wolf. (21) 


The following experiment is presented to show the results 
obtained by another method of attack. An attempt was made 
on a small isolated citrus planting at Singalong, Manila, entirely 
to remove all leaf, twig, and branch cankers, thus eliminating 
all sources of reinfection, in the endeavor to determine whether 
or not eradication of citrus canker could be effected without 
the total destruction of trees. 

The planting consisted of two rows of nursery trees which had 
been allowed to mature in place, without removal to the orchard. 
The lime trees were from four to five years old at the beginning 
of the experiment ; some of the mandarin trees were three years 
old, while others of the same species, and the calamondin trees, 
were one year old. This planting was fairly well isolated from 
other sources of infection. The trees were cultivated, irrigated, 
and cared for entirely according to the usual American orchard 
practices. The treatment of this planting was as follows : 

August 19, 1917. Formalin 1-80 solution. 

September 1, 1917. Formalin 1-80 solution. 

September 10, 1917. Formalin 1-80 solution. 

September 16, 1917. Trees pruned for the removal of twig cankers. 

September 17, 1917. 4-4-50 Bordeaux mixture plus resin, sal-soda 

October 9, 1917. 4-4-50 Bordeaux mixture plus resin, sal-soda sticker. 

November 9, 1917. Neutral Bordeaux mixture plus powdered, neu- 
tral, lead arsenate to make 1-50 mixture. 

December 17, 1917. Neutral Bordeaux mixture plus cresol-kerosene 

19,2 Lee: Citrus-canker Control 165 

December 29, 1917. Yothers's oil emulsion plus formalin 1-100. 
February 6, 1918. Cresol-kerosene emulsion. 
May 20, 1918. Neutral Bordeaux mixture. 
June 10, 1918. Neutral Bordeaux mixture. 

July 6, 1918. Neutral Bordeaux mixture plus cresol-kerosene emul- 

The plan of this spray campaign was to use several applica- 
tions of strong formalin solution as clean-up sprays at first, in 
order to reduce the amounts of citrus canker infection and so 
lessen the work of pruning out the cankered twigs and foliage. 
The trees were in many cases partially defoliated by these strong 
formalin sprays, but for the most part the cankered leaves were 
the ones to fall. The formalin sprays were followed by an ex- 
tremely careful pruning out of all twig cankers. Having at- 
tempted entirely to remove all the sources of infection in this 
way, prevention of new infection was attempted by the applica- 
tion of Bordeaux mixture. No spraying of the soil was carried 
on in as much as it had been shown by the present writer (10) 
that the canker bacteria are quickly killed out in Philippine 
orchard soils. The character of the trees and the results of 
these treatments are shown in Table 14. 

From the viewpoint of total eradication it is to be seen that this 
experiment yielded negative results. After the initial careful 
removal of all sources of infection there were recurrences of 
canker in a number of cases. In the small planting employed 
in this experiment such recurrences were noted promptly and 
removed ; in a large commercial planting of susceptible trees such 
close observation could not be maintained with financial profit. 

In this connection, however, consideration should also be given 
to the susceptibility of the host under treatment. Thus, although 
it is evident that such attempts at eradication were unsuccessful 
on the very susceptible lime varieties, a very exhaustive experi- 
ment would be necessary upon a planting exclusively of trees 
of the mandarin orange or Satsuma orange varieties before a 
similar conclusion could be adopted for such a host. 

To avoid confusion it should be made clear that, although 
calamondin and mandarin orange trees when mature are consid- 
ered resistant to citrus canker, they will exhibit a considerably 
greater degree of susceptibility when young in nursery rows. 
This has been pointed out more in detail previously by the pres- 
ent writer. (12) 


The Philippine Journal of Science 






























5^ O O 00 
<4) M CM 1-1 

oooooo ooooo 

O O iH lO 


« o 



Lee: Citrus-canker Control 


ooooooooo oooooo 

ooooooooo ooooooo 




ooooooooo o ■ ioooo 

eo o I us o 

O I o o 


S !^g !<=<=*§§§§ 


C C3 

o .5 

i § 


168 ^^^ Philippine Journal of Science 1921 


In the Philippines it would seem to be possible, by the selec- 
tion of the proper varieties and the use of the methods described 
previously, to raise citrus fruits free from citrus canker. How- 
ever, in other countries, as in Florida, where the susceptible 
grapefruit and lime varieties are already planted and very 
largely grown, control of such very susceptible varieties by the 
methods employed here would be very expensive and apparently 
not economically feasible. It seems safe to conclude that the 
practical control or the prevention of citrus canker by the pres- 
ent known methods on very susceptible hosts, such as the 
American-grown grapefruit varieties and the West Indian limes, 
can only be accomplished by the total exclusion of the disease 
from entire localities; this conclusion possibly would not apply 
to arid or semiarid countries. An alternative would be the 
securing of resistant varieties of these species, either by breed- 
ing or from among the little-known, already existing varieties 
of the East Indies and Southern Asia.^ 

In regions of the United States where the very susceptible 
lime and grapefruit varieties are not grown, control upon the 
less susceptible varieties by these methods would be possible, but 
the economic feasibility of these methods is not yet proven. In 
some regions, as in Alabama and Mississippi, where only slightly 
susceptible varieties such as the Satsuma strains are grown com- 
mercially, control methods would be practicable or even unnec- 


1. The development of the experiments has shown that in 
addition to preventive sprays the following factors contributed 
very largely to minimizing canker infection : Removal of sources 
of infection by pruning and drastic ''clean-up'' sprays ; stimula- 
tion of foliage growth to occur at periods of the year unfavorable 
to canker dissemination or development; the control of violent 
winds by windbreaks and orchard situation; and the control of 
chewing insects. 

^ The writer has obtained pummelos in Hongkong, imported from Saigon 
and Bangkok, which were entirely free from citrus canker; they were 
commercially seedless and possessed the desirable flavor and texture of 
the American-grown grapefruit. 

19,2 Lee: Citrus-canker Control 169 

2. The preventive sprays used v^ith any degree of success 
were Bordeaux mixtures and Burgundy mixtures of various 
concentrations. Lime sulphur, formalin, and ammoniacal cop- 
per carbonate solutions were unsuccessful under Philippine 
rainy season conditions. Copper sprays were not wholly suc- 
cessful, although they effected tangible reductions in canker 

3. The work has made apparent, and it has been one of the 
principal objects in the presentation of this paper to point this 
out, that the most important consideration in discussing the feasi- 
bility of control, the seriousness of citrus canker, and such 
subjects, is a knowledge of the reaction to the disease of the 
different species and horticultural varieties. The wide range 
in the susceptibility of the Citrus species necessitates the 
separate discussion of canker-control possibilities foi* each host 
or class of hosts. 

4. The conclusion is apparently safe that control upon the 
very susceptible lime and grapefruit varieties is not economical- 
ly feasible by the methods employed. Upon the sweet oranges 
of Florida origin and such varieties of less susceptiblility, con- 
trol may be practicable; further work is now in progress upon 
this point. The sweet orange of the so-called Mediterranean 
varieties, some of the lemons, and the Unshiu orange varie- 
ties comprise a class of fruits of very moderate susceptibility; 
control apparently would be economically practicable on such 
varieties. Control was very easily obtained and was hardly 
necessary upon the mandarin orange varieties, the calamondins, 
and the citrons. Attention has been called to several exceptions 
from the extreme susceptibility of the lime and grapefruit 
varieties; such exceptions may be of value for cultivation in 
regions of universal distribution of citrus canker. 

5. An attempt to obtain complete eradication of citrus can- 
ker without the total destruction of the trees was conducted on 
a small isolated plat of calamondin, mandarin orange, and lime 
trees. From the viewpoint of total eradication, negative results 
were obtained on this plat. In this case, as in others, the results 
must be considered in connection with the Citrus species used 
as hosts in the experiment. Thus these results, although ap- 
plicable to the limes, and probably to the grapefruits, should 
not be considered applicable to the less susceptible varieties, 
such as the mandarin and Satsuma oranges, until more ex- 

170 ^^^ Philippine Journal of Science 1921 

haustive experiments have been carried on with a planting 
exclusively of such a slightly susceptible host. 

6. In a district or region in which the extremely suscep- 
tible varieties such as the limes and grapefruits constitute the 
important, commercial orchards, either complete eradication of 
the affected host plants or substitution of less susceptible hosts 
is apparently the only means of preventing severe losses from 
this disease. From the results presented in this paper such 
is not the case, however, with the less susceptible species. 


1. Bedford, Duke of, and Pickering, S. U. BordQaux mixture. 8th 

Kept. Woburn Exp. Fruit Farm (1908). 

2. Idem. Copper Fungicides. 11th Rept. Woburn Exp. Fruit Farm 


3. Bourcart, E. Insecticides, fungicides and weedkillers. Translated 

from the French by Donald Grant. Scott, Greenwood and Son, Lon- 
don (1913) 282-284. 

4. Coronas, Jose. The climate and weather of the Philippines, 1903 to 

1918. Bureau of Printing, Manila, Weather Bureau Pub. (1920) 195. 

5. DoiDGE, Ethel M. The origin and cause of citrus canker in South 

Africa. Union of S. Africa Dept. of Agr. Science Bull. 8 (1916) 17. 

6. DORYLAND, E. D. Citrus canker investigations at the Singalong Ex- 

periment Station. Philip. Agr. Rev. 9 (1916) 133-135. 

7. Idem. Effects o;f formalin-Bordeaux mixture on citrus canker. Philip. 

Agr. Rev. 10 (1917) 51-54. 

8. Kellerman, K. F. Cooperative work for eradicating citrus canker. 

Yearbook U. S. Dept. Agr. (1916) 270. 

9. Lee, Atherton. Further data on the citrus canker affection of the 

citrus species and varieties at Lamao. Philip. Agr. Rev. 11 (1918) 

10. Idem. The behaviour of the citrus canker organism in the soil. Journ. 

Agr. Res. 19 (1920) 189. 

11. Idem. Action of some fungicides on the citrus canker organism. Philip. 

Journ. Sci. 17 (1920) 325. 

12. Idem. The increase in resistance to citrus canker with the advance 

in the maturity of citrus trees. Phytopathology 11 (1921) 70. 

13. LUTMAN, B. F. The covering power of the precipitation membranes 

of Bordeaux mixture. Phytopathology 2 (1912) 32. 

14. Mackie, D. B. Observations on the distribution of citrus canker. 

Philip. Agr. Rev. 9 (1916) 278. 

15. Merrill, E. D. An Interpretation of Rumphius's Herbarium Amboi- 

nense. Bureau of Printing, Manila, Bur. Sci. Pub. 9 (1917) 296. 

16. Stevens, H. E. Citrus canker, II. Bull. Fla. Agr. Exp. Sta. 124 

(1914) 42. 

17. Idem. Citrus canker. III. Bull. Fla. Agr. Exp. Sta. 128 (1914) 19. 

19,2 Lee: Citrus-canker Control Yl\ 

18. Wester, P. J. A serious disease on citrus fruits. Philip. Agr. Rev. 7 

(1914) 423. 

19. Idem. Additional observations on the citrus fruits in the Philippines. 

Philip. Agr. Rev. 10 (1917) 105. 

20. Idem. Notes on Philippine citrus fruits. Philip. Agr. Rev. 12 (1919) 


21. Wolf, Frederick A. Citrus canker. Journ. Agr. Res. 6 (1916) 95. 

22. YOTHERS, W. W. Spraying for whiteflies in Florida. Circular U. S. 

Dept. Agr. Bur. Ent. 168 (1913). 


Plate 1 * 

Citrus leaves showing cankers following attacks of leaf-chewing insects. 

Plate 2 

Citrus leaves showing cankers following trail of leaf miners. 


Lee: Citrus-canker Control.] 

[Philip. Journ. Scl, 19, No. 2. 



Lee: Citrus-canker Control.] 

[Philip. Journ. Scl, 19, No. 2. 




By Otto Scheerer 



The people at present officially designated as Kalinga occupy 
a central position in the interior of northern Luzon. They have 
for neighbors in the north the Apayao, in the west the Itneg 
(Kal. Itnog), in the south the Ibontok, and in the east the Iba- 
nak and Gaddang. Their country consists, in the main, of a 
portion of the high ridge of the Cordillera Central and of the 
lower sierras which from this backbone stretch forth toward 
the east, sending their waters in numerous streams to the Rio 
Chico de Cagayan. On the upper course of this river is situated 
the town of Lubuagan, the capital of the political subdivision 
of the Mountain Province known as the subprovince of Kalinga. 
The accompanying sketch will show the general disposition 
of the Kalinga territory. (See page 177.) 


Individuals of this tribe spoken to by me called themselves 
Kalingga or, more idiomatically, Kalingka. Since kaling-d in 
Ibanak and kalinga in Gaddang (the latter in the spelling of 
P. Malumbres) both mean ''enemy," it may be surmised that 
the name in question was orginally bestowed upon the people 
from outside. 


In spite of the Kalinga having merited up to recent times the 
fame of bold warriors and inveterate head-hunters, their terri- 
tory appears to be almost nowhere demarcated by sharply drawn 
ethnic boundary lines such as those antecedents might lead us 

' The texts here published are the first coherent records made of the 
speech of some little-known dialect groups of northern Luzon. While 
hardly more than raw material, their publication must be considered 
as opportune since, with the profound and rapid changes now going 
on in the Islands, it becomes a matter of doubt if dialects like the present 
will still be in existence when a systematic investigation may become 
possible at some time in the future. 


176 The Philippine Journal of Science 1921 

to expect. Almost on all sides zones of transition are reported, 
peopled by the offspring of the Kalinga's intermarriage with 
his neighbors. It becomes thus difficult to treat of the Kallnga 
as of a unit without making almost at every step allowance for 
intermixture. Modern ethnographers like Worcester, Beyer, 
and Cole, who, after personal investigations on the spot, have 
first recognized a distinct and prominent tribal unit among the 
jumble of so-called "tribes" reported in Spanish times from the 
territory above defined, and who have fixed on that unit the 
name Kalinga, have found it necessary to subdivide this people 
into a number of groups according to admixture of blood, cul- 
ture, or dialect.^ Beyer distinguishes the following groups: 

1. The pagan Gaddang. 

2. The Kalaua or Kalagua. 

3. The true Kalinga of the lower Saltan, Nabayugan, Bukao, and Ta- 

lifugu river valleys. 

4. The Balbalasang-Ginaang group. 

5. The Lubuagan-Sumadel group. 

6. The Mangali-Lubo group. 

Of these he says that in the present state of our knowledge 
this must be considered as a rough and tentative grouping 

The following items of information gathered by me from 
m.embers of the Balbalasang-Ginaang group support the theory 
of a close relationship existing among the different Kalinga 
dialects. The Balbalasang people designate their speech as 
kainalingka> (Salegseg Kenalingke,) At those occasions when 
the chiefs of the various townships constituting the subprovince 
of Kalinga meet at the capital Lubuagan to confer with the 

' Cf. Worcester, The non-Christian tribes of northern Luzon, Philip. 
Joum. Sci. 1 (1906) 791-805 and 818-826; also Fay-Cooper Cole, Distri- 
bution of the non-Christian tribes of northwestern Luzon, Am. Anthro- 
pologist 2' (1909). An older conception of the Kalinga is found in the 
following translated quotation from Blumentritt, Versuch einer Ethno- 
graphic der Philippinen. Gotha (1882) 36: ''According to Semper (Erd- 
kunde, X, 256) the name Calinga seems to be a collective designation 
of unknown signification since thus are called also all the pagans who 
inhabit the provinces of Isabela, Cagayan, and Nueva Vizcaya. I desig- 
nate here with this name that pagan Malayan tribe which lives in the 
same mountain stock as the Aripa, though only in its northern part, and 
goes especially by the name Calinga. But little is known of them.'* 

* Beyer, H. Otley, Population of the Philippine Islands in 1916. Manila 
(1917) 50-51. On page 43, in treating of the Ibanak, the author says: 
"At least one quite different dialect exists. This is known as the Itavi 
or Malaueg. The people speaking it are probably christianized Kalingas." 


Scheerer: Texts from Balbaldsang-Gindang 


Lieutenant-Governor, they and their followers speak among 
themselves kainalingka, in which they are able, according to 
my informants, to make themselves mutually understood, with 
more or less difficulty, in spite of dialectic differences. Two 

Fig. 1. The subprovince of Kalinga. 

townships alone are pointed out as not coming within this range 
of general intelligibility; those are Bakali {1<iha kali other 
speech) and Kalakkad; to communicate with these the Iloko 
language is made use of.* 

*The map of Northern Luzon of the Coast and Geodetic Survey (1912), 
on which my sketch is in part based, shows on the upper course of what 
is called farther down the Tardi River the two settlements of Bocale 
and Calacad, situated east of the Mangali-Lubo district. To Governor 
Carpenter, of the Department of Mindanao and Sulu and director of the 
Bureau of Non-Christian Tribes, I owe the information that by Executive 
Order No. 53 of May 29, 1914, "all that territory lying south of the 
Tardi River and the territory drained by the Siffu River now situate 
in the subprovince of Kalinga, Mountain Province," was transferred from 
the said subprovince to that of Bontoc. 

180732 4 

178 l^he Philippine Journal of Science 1921 

In view of these facts the existence within the Kalinga terri- 
tory of a goodly number of dialects may be expected. Whether 
these, once known in their totality and compared with one 
another and with the surrounding stocks, will show common 
charateristics leading to the establishment of a typical Kalinga 
speech as represented in its purest form by one or the other 
of these dialects, is one of the interesting questions the solution 
of which will reward the linguistic explorer of this region. 
The present state of our knowledge in this regard is charac- 
terized by Beyer in these words : 

The mixture of Kalinga dialects is as confusing as their type and 
culture, and in passing from one district to another most striking dif- 
ferences in phonetics are observed. Structurally those of the west seem 
to resemble the Iloko group, while the eastern dialects are closer to 
the Ibanag. We have too little information at the present time to attempt 
a classification of the Kalinga dialects, or even to state their number.*^ * 

The grouping of these dialects thus will have to follow for 
the present merely geographical lines. 


The only published records known to me of the speech of 
people coming under the designation of Kalinga as now in use 
are the following : 

1. Schadenberg, Alexander, in Beitrage zur Kenntniss der Banao-Leute 

und der Guinanen, Gran Cordillera Central, etc., Verhandl. Berl. 
Anthr. Ges. (1887) 152-159. A vocabulary of some 660 words 
of the '^Guinaan dialect as spoken in the rancheria of Copacopa." 
In this as in another vocabulary from central northern Luzon, 
this earliest scientific explorer of those regions shows himself to 
have fallen only too often a victim to the habit of those natives 
of substituting Iloko or other lowland terms for their own idioms 
when dealing with foreigners. 

2. Meyer, Hans, Guinanisch-Tinguianisches Vocabular, in Eine Welt- 

reise, Anhang: Die Igorroten. Leipzig (1884). Some 120 words. 

3. Scheerer, O., in Linguistic travelling notes from Cagayan, Anthropos 

4 (1909) 3 and 4. A list of QQ words from the rancheria of 
Gobgob on the Rio Chico. 


The group of Kalinga whose speech is recorded in the follow- 
ing texts is at home on and around the upper course of the 
Saltan River where, according to Cole, they have married with 
the Igorot and Tinggian. 

This author says: 

^ Beyer, op. cit. 51. 

19,2 Scheerer: Texts from Balbaldsang-Gindang 179 

The towns of the upper Saltan river have drawn much from the three 
tribes which have contributed to their population, but the Tinguian mate- 
rial culture is the most pronounced. The typical costumes, method of 
hair-dressing and the arm-beads of the women, in vogue in Abra, are 
all found here. Agriculture is extensive, and the terraced fields compare 
favorably with those of Bontok. All kinds of domestic animals known to 
the natives of the coast are possessed by these people. The best iron- 
work of northern Luzon comes from this section, and their head-axes and 
spears have a wide distribution over the whole Tinguian and Kalinga 

Beyer says of them : 

In physical type they are more or less like the true Kalingas, but 
their culture and speech is a blended mixture of Apayao, Kalinga, Ting- 
gian, and Bontok. Out of this mixture there have been a number of 
curious and unique cultural developments.'^ 

I may add to this that the different members of this group 
who were spoken to by me called themselves, as already stated, 
Kalingkd, giving therewith evidence of their being conscious of 
belonging all to one and the same tribe. Since olden times they 
maintain trade relations with the Tinggian in the west, whom 
they call Itnog and whom they consider their friends; going as 
far as Banged (Span. Bangued), they take down to them rice 
(Bal. pdkoi unhulled rice, phindyu hulled rice) and tobacco (Bal. 
taphidko) in change for salt (Bal. asin), pigs (Bal. Pfuyok^), 
woven stuffs (Bal. lopot), beads (Bal. ullayau), metal pipes 
(Bal. siwako pipe), etc.; since the advent of the American 
regime they also trade with the people of Tuao and Piat in 
Cagayan to whom they bring likewise rice in return for salt, 
iron (Bal. Phydyang), etc. Their main trade route lies along 
the Saltan River (Sal. SaKtan) whose banks, hemmed in by 
steep rocks forming narrow and winding gorges, have given it, 
in local parlance, the name MaKokpfub, the cliff-bound. The sig- 
nificance of this river as a landmark is evidenced by the fact 
that the terms chaya and Idkud, denoting two cardinal points, 
have for the river-dwellers primarily the meaning of "up river" 
and "down river," respectively, and only according to the orien- 
tation of the river those of "west" and "east," respectively. 


Balbalasang (locally Phyaiphyaydsang, from phyaydsang 
maiden) is officially stated to count at present approximately 

"Op. cit. 342. 

'Op. cit. 50. 

"O quite open, as generally in these dialects. 

180 ^'fee Philippine Journal of Science 1921 

1,048 souls. It is to be conceived as a district comprehending, 
besides Balbalasang proper, a number of "rancher ias'' or set- 
tlements (phophoyoi) . The following were indicated to me as 
belonging socially, though not necessarily politically, to Bal- 
balasang, and as speaking with little difference the same dialect : 

Pasual. Kallagan. Talagan. Malibkong. 

Pattekyan. Saltan. Photlok. Phulayayau. " 

Inilagan. Sisik-an. Phyanagan. 

The chief peculiarity of pronunciation that strikes the ear 
of the traveler who enters the Balbalasang district from the 
west is the sound which I represent by ph and which before a 
becomes phy. From a comparison of the word phyaydsang 
above quoted with Ilk, baldsang maiden it may be seen that ph 
replaces other Phil. 6, and that the second y in that word stands 
for other Phil. I. Similar examples are: Ilk. bato stone, Bal. 
phyato; Tag. guhat forest, Bal. kinophyat same meaning; Ting. 
banug a certain eagle, Bal. phyanug. The sound here in ques- 
tion is a pure bilabial, transitional from a stop to a fricative, 
and in so far difficult to judge correctly as not only its position 
in the word and the sounds preceding and following it are apt 
to make it fluctuate from p over pf to /, but also different 
speakers do not all produce it in exactly the same way.^° On 
the question whether it is in all cases unvoiced — as I give to 
understand by the uniform writing ph — I must reserve a judg- 
ment which could be given only after a more extended investi- 
gation than was possible to me. I may say, however, that before 
u this sound approaches either pf or / so closely that I was 
tempted to write, for instance, uppfu (young of animal) for 
upphii, nadfus (finished) for nadphus, and others similarly. 

This fluctuating sound is also found in the dialects of Ginaang 
and Salegseg where it shows again some modifications. 

Another notable feature of Bal. phonology, found likewise 
in Gin. and Sal., is a velar stop lying, in strength, between g 

'Names here given are as locally pronounced; outsiders pronounce 
them somewhat differently. 

^^ Though different in origin, this sound appears to be very similar to 
the one described! by Conant in his F and V in Philippine languages, Bu. 
Sci. Div. Ethn. Publ. 5' (1908) 139; treating of the faulty pronuncia- 
tion of p by individual Filipino children from various tribes, the author 
says that "the sound is produced by expelling the air through the lips 
when closed, but so relaxed that a very slight explosion is immediately 
followed by an almost imperceptible spirant, the result being a pf sound, 
the constituent elements of which are so blended as to be hardly dis- 
tinguishable from each other." 

19.2 Scheerer: Texts from Balbaldsang-Gindang 181 

and k. Also respecting this sound a definite statement as to 
the presence of voice must be reserved to further more leisurely 
observation. The general impression was that of a strongly 
articulated g, my unwillingness to set this sound down as a fc 
being perhaps due to the fact that it is produced considerably 
farther back than k in English. Before a, especially before 
stressed a, it acquires the same strength as, for instance, c in 
English cat, even with some aspiration following it: kaling-kd 
almost kaling-khd. Disregarding fluctuations observed, I rep- 
resent it for the present uniformly by k : tdku person, iningkdu 
there was. 

The consonantal diphthong ch, evolved, as in Inibaloi, from 
d, and which occurs in all three dialects here recorded, was pro- 
nounced much softer than in the first-named language. A study 
on the spot might lead to the adoption of a symbol corresponding 
to dj of the International Phonetic Association. 


Salegseg (locally SaAogsog) is reached from Balbalasang in, 
four hours on horseback down the Saltan River. Official sta- 
tistics give the population as about 940. Affiliated settle- 
ments speaking practically the same dialect are: 

Pfu^o (Bolo). Chuso(k). Koewoean. Possa. Pottau. 

Upoif. Ka-wong. Ota. Legleg. Kilayun. 

Lopw6ng. Alengngag. Nawoi. Ta-wang. PfuayantoiC. 

My informant, a native of Salegseg proper, spontaneously 
stated that their dialect was about the most difficult to under- 
stand for the Kalinga of other districts, and this on account of 
its peculiar phonetics. Asked for the name of their speech, he 
unhesitatingly gave it as kenalingkd (<kainalingka) . The chief 
phonetic peculiarities are: 

1. A mixed vowel of the o class replacing other Phil, a according to 

rules yet to be established. Being quite open, it is best repre- 
sented by the oe of the International Phonetic Association. Isin. 
gawa middle, Bal. kawa, Sal. koewoe. Occasionally this sound be- 
comes rather indistinct (the 9 of the Int. Phon. Ass. ) : Tag. buaya 
alligator, Bal. phudya, Sal. pfuceya; Ilk. dua two, Bal. chua, Sal. chu9, 

2. The peculiar Bal. bilabial ph, representative of other Phil, b is 

found in Sal. in about the same average form only before i: Tag. 
gabi night, Bal. laphi, Sal. laphi. Before an original a it becomes 
what might be called a hard b (I employ for it the symbol j) 
and thus forms— seemingly according to the quantity of the syllable- 
either the syllable dee (Inib. mabayag what lasts much time, Sal. 
madcey9g; Span, bayoneta bayonet, Sal. dceyoneta; Ilk. bdsa idea 
of reading, Sal. dcessa; Ilk. ibagd what is told, Bal. iphyakd, 
Sal. idoege), or the group diia (Phil, bato stone, Sal. d'iiat6; 

182 The Philippine Journal of Science 1921 

Phil, hasd wet, Sal. duassd). An exception would seem to be 
Sal. do^oi house (Bal. phoyoiy Ilk. baldi). The essential dis- 
tinction between the sounds represented by b and p is, of course, 
the presence or absence of voice, a question which, in reference 
to the sound here under review, must be left for future investi- 
gation. Before u (o) a fairly distinct, though very light, affricate 
is heard: pfuoeys alligator, quoted above; Ilk. humangon one rising, 
Sal. pfumangon; Ilk. umisbd who makes water, Sal. umispfo; Phil. 
Bontoky Sal. Pfuntok, 
3. A sound produced by articulating an I with the tip of the tongue 
not touching the alveoli but passing more or less freely through 
the teeth, as my informants very clearly demonstrated to me. 
I use for it the symbol R. The attempt to produce this sound, 
which is not found in English, is best made with an initial vowel; 
thus with o, for instance, a diphthong approaching oi will result; 
compare Ilk. luppo thigh, Inib. ulpo Sal. oKpo, The name of the 
river Saltan, of the town Salegseg, and of the barrio (detached 
suburb) Bolo are locally pronounced SaS-tan, Sa/iogsog, and PfuS.o, 
respectively. The Spanish word for highroad, calzada^ becomes 
in Sal. ka/isa, the final syllable da being left out presumably 
under the mistaken belief that it is the possessive suffix of the 
third person plural da. Also in Bal. the change from other Phil. 
I to if will be found through comparison (Ilk. dakkel large, Bal* 
chakkoi, Sal. chakkoK) ; the pronunciation of the sound is there, 
however, so glib and free of an admixture of I that I have not 
considered it necessary to use a special symbol for it. In Sal. 
the sound in question is often very deceptive and in the foot- 
notes to the texts there will be found pointed out cases where it 
sounds almost like I. 

Mention is to be made here also of the fact that tenues in 
final position are pronounced, especially in isolated words, with 
extreme softness making them sometimes scarcely audible : Chii- 
so (k) a part of Salegseg, dwa(k) waist, choewaik) a semireli- 
gious ceremony (Ting, dawak), awi(t) load, nasH-yo (p) one 
sleeping, etc. Apparently this is due to the release of the stop 
being effected in a scarcely perceptible manner. 

Ginaang, a town situated about half a day's ride southeast 
from Balbalasang, and about two hours northwest from Lubua- 
gan, between the Rivers Pasil and Tabia. The inhabitants, said 
to have been in former times the terror of their more peaceful 
neighbors and referred to by the Spaniards as **la tribu de los 
Guinaanes," number according to recent official statistics about 
1,051 souls. Rancherias popularly affiliated with Ginaang and 
speaking, with perhaps slight difference, the same dialect, are: 







19,2 Scheerer: Texts from Balbaldsang-Gindang 183 

The dialect of Ginaang, as recorded in the following texts, is 
quite evidently part of the general Kainalingka speech-group; 
one of my informants, however, a son of the old fighting chief 
Atumpa of Ginaang, had for it the special name Ginindang and 
disapproved of the generalizing term Kainalingka because of 
the existing dialectic differences. He had to admit, nevertheless, 
the similarity of their dialect to that of Balbalasang and, more 
especially, to that of Balatok, another considerable town some 
distance up the Pasil River, with whose inhabitants they have 
no difficulty to converse. 

Ginaang phonetics, while in general very similar to those of 
Balbalasang, show in certain points a distinct weakening. The 
characteristic ph sound has less friction and approaches more 
a strong 6. Similarly the dubious velar stop mentioned under 
Balbalasang inclines in Ginaang more to g, so that words like 
tdku, ingkdu, and others I was tempted to write tdgu, inggdu, 
etc. The consonantal diphthong represented in English by ch — 
and thus given also in the texts — often sounds rather like d3 
than tj, while some words seemingly ending in d, as maid, ukudy 
really wind up with a very slender hesitating sibilant: maid-\ 
ukud-\. The almost inaudible final k mentioned under Salegseg 
is found also in Ginaang where, moreover, initial and intervo- 
calic Phil, k are elided: Inib. kalbian last night. Gin. ayabian; 
Bal. kdisan gone away. Gin. disan; Inib. karagwian custom, 
Gin. achawian; Sal. kan seka to thee. Gin. an si-a; Ilk. baldi ko 
my house, Sal. phoyoi-o. In place of the disappeared k no glot- 
tal check could be detected, nor were my informants conscious 
of any such ; all that remained was a slight hiatus.^^ 

There remains finally to be mentioned a quality which in all 
three dialects consonants occasionally revealed to possess for 
binding syllables together. Thus a word which was heard pro- 
nounced, in the ordinary flow of speech, imuson (object of 
asking), showed, when spelled by the speaker in syllables, the 
division i-mits-s6n, the s belonging as much to the second syllable 
as to the third, without becoming thereby audibly geminated. 

" The regular elision in the cases above cited of k of other Philippine 
languages clearly indicates the distinction to be made between this original 
k and the dubious k before mentioned which has evidently been evolved 
from g: for example, ug-ukud, story, shows g closing the first syllable 
and in its stead k in intervocalic position; Ilk. kigau becomes in Bal. 
kikau; compare also words ending in g and followed by connective a: Bal. 
kaag plus connective a gives kaag-ka, monkeys — which (Ilk. kaag the 
young of the monkey) . Note also that in Gin. taphyako, a loan-word from 
Span. tabakOf k was not elided. 

184 The Philippine Journal of Science 1921 

(Wherever a long consonant was sounded this was recorded in 
writing by doubling it: ummoi, pillak.) There was no oppor- 
tunity to make this point an object of special study; it may, 
however, be kept in mind when dealing with the coalescence 
which is shown in the texts often to take place between the 
connective particles a, ot, and the final consonant of the preced- 
ing word: chakkoi-ya phyanug large eagle, akit-ta iphil little 
crying, phyanug-ka mandnap eagle-which searched, phyato-wot 
stone-and, liang-ngot cave-and. The language did not, however, 
seem to be very strict in this respect, for, while the borrowed 
consonant was at times pronounced even after quite a consider- 
able intervening pause, leading me at first to the belief in the 
presence of such independent words as ta, toot, at others there 
was either a simple addition of the particle to the preceding 
word (achdtjaoma Hang deep cave), or the consonantal accretion 
before the connective was lost through a pause. 

The coalescence just treated finds its reverse in a number of 
words the staccato pronunciation of which leads to such writing 
as: man-di-dyam, na-dm-dmod, indm-amdan, pit-ongona, inum- 
ummoi, sdg-on, man-iwd-aA and others. 

In words not marked with an accent the weight of the pro- 
nunciation falls as a rule on the penult. 


The following texts were collected by me during vacation 
time at the Teachers' Camp in Baguio and at the farm school 
at Trinidad, Mountain Province, from some Kalinga students 
of about twenty years of age. Thanks to the courtesy of the 
authorities concerned ^^ a series of sessions could be arranged 
with these young men. It was at first tried to have the students 
write down their stories, an attempt which proved, however, 
an entire failure, as the boys either frankly stated their inability 
properly to represent in writing the uncommon sounds peculiar 
to these dialects, or rendered sounds and words in a way that 
made it difficult for them to decipher after a while what they 
had written. I had thus to undertake the not easy task of fol- 
lowing the flow of the story-teller's relation while trying, at the 
same time, to put into writing his utterances which at the time 
being were, of course, absolutely unintelligible to me. Still, 
after repeated readings and corrections each story took ulti- 

" The facilities for my work kindly granted by Professor Wright, di- 
rector of the Trinidad Farm School, and by Mr. McCann, of Teachers' 
Camp, Baguio, are sincerely appreciated. 

19.2 « Scheerer: Texts from Balbaldsang-Gindang 185 

mately a form that was pronounced correct, whereupon the 
translation into English was taken up, and one or the other 
point of grammar discussed. 

The stories thus recorded were: 


1. Sacha phyanug kan aphit; The eagle and the child; from the town 

of Balbalasang (Phyaiphyayasang). 

2. Sacha Changatag kan cha kaag; Changatag and the monkeys; from 

the rancheria of Phulayayau. 
8. Sachat Iphudyan nangdyau ud Pattekyan; Those from Buaya go 
on a headhunt to Patikian; from the rancheria of Pattekyan. 


4. Se AKukdna kennan chit pfuceys; Si Alugan bitten by an alligator; 

from the rancheria of PfuiCo. ^ 

5. Ug-ukud kan Lumawig; The story of Lumawig; from the rancheria 

of Ka-wong. 


6. Ug-ukild chat man-aman nagchaydn; The story of the father and 

child buried by a landslide; from the town of Ginaang. 

To these texts is added: 



7. Mepanggep ta Itneg di Abra; About the Itneg of Abra. 

The last account, which was obtained from a young Tinggian 
student of the farm school already mentioned and is to my 
knowledge the first text ever published from that language, may 
afford students an opportunity to compare the Tinggian with 
the Balbalasang-Ginaang dialects which latter are said to have 
been influenced by their western neighbors. 

As regards the translation of the texts I should say that the 
liberties which I have occasionally taken with the English lan- 
guage are to be explained by the desire to reflect, as much as it 
can at all be done, the peculiarities of Kalingga diction. 

A list of the personal and possessive pronouns in all dialects 
here recorded is added at the end. 



1. Sachit nasulit sin phopho- 1. Long ago near the town 

yoi ud Phyanakan iningkau of Banagan there lived a large 

chakkoi-ya phyanug. Inaikau- eagle. Every day it came to 

wa umoi sit phophoyoi mang- the town to catch pigs and 


The Philippine Journal of Science 


ayas phoyok ya kikau. Maid 
makapatoi sit phyaniig tei 
umogyat chachit taku. 

2. Sat osan kidkicham ining- 
kau osan aphit-ta man-ai-ayam 
sit pawai. Man-lin-linawa sit 

3. Masulit man inilan chit 
phyaniig chit aphit. Manip- 
pokoi chit phyaniig ot chapyo- 
tonat aphit ot itaudna -uschit 

4. Mangkolis mat aphit na- 
aphiis ngummato chit phya^t 
nug. Na-am-amod chit ogyat 
chat taku sit o6t-tap6ncha man 
chit phyanug-ka nang-aya sit 
aphit innokna uschit liang. 

5. Unuchon chat taku ot mi- 
chatongcha man sit philig; 
maid maoi tai chopiyas on na- 
taknang. Chuchongyoncha chit 
iphil chit aphit ngato umya 
akit chit chongyoncha. Masu- 
lit man chit manta-tangachan- 
cha, maid chongyonchas iphil 
na. Mampaiilicha tai pasikon- 
chan kinan chit phyaniig. 

calves. There was nobody 
who could kill the eagle for 
the people were afraid. 

2. On one afternoon there 
was a small child playing in 
the front-yard of a house. A 
delightful cool air was at the 
place where it played. 

3. After a while the eagle 
saw the child in the yard. 
The eagle swooped down and 
snatched the child and flew 
away with it to the mountain. 

4. When the child screamed, 
the eagle had already risen 
high. Excessive was the fright 
of the people when looking up 
at the eagle that had taken the 
child and carried it to the cave. 

5. The people followed it and 
reached the mountain; it was 
impassable for it was very 
steep and high. They listened 
for the crying of the child 
above, but little was heard by 
them. After looking up for a 
while they did not hear its 
crying. They went back for 
they thought it had been eaten 
by the eagle. 

1. Phophoyoi village, group of houses; from phoyoi house. 
iningkdUy past of ingkdu, finding one*s self, staying, being. 
phyanug probably the large, monkey-eating eagle, Pithecophaga jefferyi 
Grant, found on Luzon, Samar, and Mindanao. The Ibaloi of Ben- 
guet have a similar story of this bird called by them saragmd and 

.found there only in the foothills. 
mang-dyas from mang-dya one getting+s. 

2. Aphit small child, baby. 

8. man (probably ma-\-n, compare later mat from ma-\-t) seems to mark 
the progress of events. 
chapyotonat from chapyoton object of snatching+na its-f-article t, 
4. nadphus expresses finished action. 

ngummato, past of ngumato from ngato above-}- infix um, 
o6t-tap6n; imper. otapom let (it) be the object of thy looking up; 
past intap. 

19, 2 

Scheerer: Texts from Balbaldsang-Gindang 


6. Mawakas ot umoi chit 
amana mantangad manum- 
phiis ; nagngoi nat akit-ta iphil 
chit anakna. Namingsana 
chingngoina. Nampauli chit 
amana tai achina naka-aya sit 
anakna. Chakkoi-ya phyaphya- 
winan chit kaisan tai adsana 
na-ila chit anakna. 

7. Maphitil man chit aphit 
maid mangai-anas kanona tai 
maid apoi ya maid phinayu. 
Maphitil mana-iinai, nangan 
sichat inayan chit phyaniig-ka 
ugsa onno phoyok. 

8. Makaduan phuyan ot 
chummakkoi chit aphit ot 
chummachakkoi pai chat up- 
phun chit phyamig. Oyog man 
chat upphun tumaud chinom- 
chom chit aphit on man-ayan. 

9. Ot osan phigphikat man 
kaisan chit inan chat upphQ. 
Postona chit ikin chat upphii 
ot ilayugna chicha. Ot inam- 
ammaam chit aphit-ta niphia- 
nat sit pita. Ichayana chat 
chuan upphii sit phoph6yoi. 

6. The following day the fa- 
ther went to look up again ; he 
believed to hear some little cry- 
ing of his child. He heard it 
only once. The father went 
back for he could not take his 
child. Great was his sorrow 
on going away for he had not 
found his child. 

7. The child was hungry, but 
there was no getting any food 
because there was no fire and 
no rice. Being very hungry it 
ate from the prey of the eagle 
which was deer or pig. 

8. After two months had 
passed the child had grown big- 
ger and also the young of the 
eagle had grown up. As the 
young eagles were able to fly 
the child thought of going 

9. Then, one morning, the 
mother of the young eagles 
flew away. The child tied the 
legs of the young to use these 
for descending to the ground. 
And slowly the child sank down 
to the earth. It took along the 
two young eagles in going back 

to the village. 

5. unuchon object of following. 
michatongy possibly from *7naichat6ng. 
maid mdoi not passable. 

chuchongyoriy probably a progressive form of * chongoyon obj. of lis- 
pasikon object of thinking; past pinasig, 
kinan, past of kanon object of eating. 

6. umoi one going; we would expect the past ummoi, but this tense appears 

to be far from being regularly employed. 
manumphus: our adverb "again" appears as a verbal form auxiliary 

to mantangad, 
nagngoi a form said to denote one who believed to hear. 

7. inayan what was taken (by the eagle). 

8. oyog seems to express the idea of enough, sufficient. 


The Philippine Journal of Science 


10. Sumaai man chit aphifc 
nascha-au chat taku tai pasi- 
konchan natoi. Natayok chat 
amana ya susunudna tai su- 

11. Mangwa man chat taku 
nangwachas silib. Iphya-ud- 
cha chat chuan upphu sit pa- 
wai ot sukaoncha. Masulit 
manchi iningkau chingngoichan 
kas phyalin nampaipo sit kino- 
fyat. Chumakko-chakkoi chit 
anunong an chingngoi chat ta- 
ku ot lumokwas man si phya- 
niigka mananap sichat upphu- 

12. Mailana man chat up- 
phuna manippokoi-yot chapiyo- 
tona chat upphuna sit suka ot 
manokuyoscha chit taku ot 
patayoncha. Insangapatoicha 
chat chuan upphuna. Maa- 
phus chiyot maichon ogyatan 
chat taku. Naaphus. 




1. Sat osan ai-aikau Si Chan- 1. On a certain day Si Chan- 
gatag ud Saitan ummoi ud gatag of Saitan went to sell 

9. kaisan was said to be the past of manayan one going away. 
poston object of tying, past pinsot. 

ildyug with the help of which one descends; instrument of descending. 
inam-ammaauy an auxiliary verbal form expressing the idea of '^slowly" 
and being in the same (past) tense as the principal verb niphianat. 
ichdyan (probably from chdyan road, trail) what is taken along. 

11. mangwa includes such meanings as: make, think, say, decide, etc.; 

past nangwd (contained in nangwachas). 
iphyd'Ud, cf. Ilk. ihalud, 

12. maichon, from maid there is not, and suffix, on, indicating direct object. 

10. When the child arrived 
at home, the people were star- 
tled for they had thought it 
dead. Its parents and bro- 
thers and sisters rejoiced over 
its arrival. 

11. The people decided to use 
a trick. They bound the two 
eaglets on the yard and planted 
sharpened sticks in the ground 
around them. They had been 
there a while when something 
like the rushing of wind was 
heard by them which came from 
the forest. Louder grew the 
sound which the people heard, 
and what appeared was the 
eagle looking for its young. 

12. When it saw its young, 
it shot down to snatch them 
from among the sharpened 
sticks, and then the people 
rushed to the place and killed 
it. Together with it they 
killed the two eaglets. And 
thus was ended at last the fear 
of the people. Finished. 



19, 2 

Scheerer: Texts from Balbaldsang-Gindang 


nanlako si kusi. Kinimatana 
chit chuan phokoi-ya kusi on 
ina ingina si luwang lischin 
phophoyoi ud Photlok. 

2. Ingkau man sichit kawan 
chit kiniifyat iningkau chat 
kaag sit chayan. Ot umillong 
yan tai naphyanikoi sit ining- 
kawan chat kaag on inilana. 

3. Masulit manot ayana chit 
phyato-wot pit-ongonat osan 
kaag. Masulit manot umachu 
cha chit kaag sit kayu on man- 
ischung kan sia. 

4. Paiyan ayanat osan phya- 
to-wot pit-ongona c h i c h a. 
Auni manot lumsa chat chuan 
kakuiiphyanchan ummoi kan 
sia. Sinongpatna chicha. 

5. Maphyayag manot naamin 
nilumsa chat iningkau sit kayu 
on manokuyus kan sia ot malig- 

6. Iningkau ud achayoma 
liang uschit ikid chit chayan. 
Tinainana chit awit nan kiisi- 
yot phumtik sit liang-ngot 

jars. He carried suspended 
from a pole on his shoulder a 
couple of jars which he went 
to sell for a water-buffalo in 
the town of Photlok. 

2. When he was in the middle 
of the forest there were mon- 
keys on the road. And he just 
sat down for rest as he was 
tired, at the place where there 
were the monkeys he had seen. 

3. After a while he took a 
stone and threw it at a mon- 
key killing it. Thereupon the 
monkeys became many on the 
trees and they peered down 
upon him wistfully. 

4. Again he took a stone and 
stoned them. After that there 
came down two of their leaders 
who went at him. He cut them 

5. After a while all of them 
that were in the trees came 
down and rushed upon him and 
he became excited. 

6. There was a deep cave at 
the side of the road. He left 
the load of jars and ran to the 
cave and many were also the 

1. kinimdta; a means of carrying loads is by a pair of baskets, kimdta, 

suspended from the extremities of a pole, aseu, carried over the 

shoulder; mangimata one thus carrying a load; kimatdon what Is 

thus carried. 
phokoi a classifying auxiliary to numerals used to count globular things 

such as jars, demijohns, coconuts, etc. 
ingina si luwang; also inglnat luwang might be said; luwang is the 

nuang of other northern dialects. 

2. kawa middle, Isin. gawa. 
kinufyat forest, cf. Tag. gubat. 
umillong one resting; ydn meaning "just." 

3. pit-ongonat: pit-ongon object of stoning (Past pinit-ong) +na his-f 

man-ischungy cf. Bont. umuschung one looking do-wn watching. 


The Philippine Journal of Science 


achun unai chat kaag-ka me- 
ton-ud kan sia ; umya iningkau 
phyachangnan anchu. 

7. Ipainokna chit potod na sit 
liang ; pasawayona chit awakna 
on mampatoi sichat kaag-ka 
mamatoi kan sia. Awad pai 
nilumnok sit aphut tinogmanat 

8. Achu man chit pinatoina 
on sachat kaag nangwachas si- 
lifcha. Ummoi chat lichum sit 
tongod chit liang-ngot kuyQ- 
phyanchat tongod chit phyato. 

9. Na-amin pai chat kaag sit 
sauphian chit liang nisukat 
chat sumasayum. Maphyayag 
manot sachat kaag-ka iningkau 
sit tongod chit phyato lummos- 
phucha sit chayom chit liang- 
ngot sa chiyon tiniliucha on Si 
Changatag-kot iphusnagcha ot 
mapoipoitoyancha ingkanas 
maicha unai ud maiwayang sit 

10. Sa chiyon naichon Si 
Changatag ta kinan chi kaag 
sit chayana inoina. Naaphiis. 

monkeys who followed him; 
however, he had with him his 
long bolo. 

7. He put the lower half of 
his body into the hole, stuck 
out his waist, and killed the 
monkeys who wanted to kill 
him. Whenever one entered 
the hole he cut off its head. 

8. Many were those killed by 
him when the monkeys used a 
stratagem. Some of them 
went to the back of the cave 
and dug a hole at the back of 
the rock. 

9. When the monkeys at the 
entrance of the cave were all 
exterminated many others came 
to replace them. After a while 
also those monkeys who were 
at the back of the rock passed 
through to the interior and 
thereupon caught hold of Si 
Changatag who was dragged 
out by them and torn to pieces 
until no more was left of his 

^10. Thus perished Si Chan- 
gatag bitten to death by the 
monkeys on the road which he 
had traveled. Finished. 

4. sinongpat, past of songpdtoriy object of cutting down. 

6. tinainan, past of tainan what is left. 
phumtik one running, past phinumtik, 

7. tinogmaj past of togmdon: togmdok my object of cutting off. 

8. sUifcha their trick; the last sound of silif is a very slender fricative. 

9. tiniliu, past of tiliwon what is seized. 

10. kinan meaning generally "eaten" but also "destroyed." 
inoiy past of ayon what is passed, as a road walked over. 

19, 2 

Scheerer: Texts from Balbaldsang-Gindang 





1. Sat osan ai-aikau iningkau 
chad Iphuayan nangayau ud 
Pattekyan. Awad kad chaton 
takun nampaipod Phuaya to- 
yumpuyu cha ot lima. 

2. Umachanicha man ud Pat- 
tekyan umillongcha yan ta 
mangancha ta payot naphilog- 
chan makapatoi si kancha. 

3. Umya lamong ot tai ining- 
kau osan layakin Ipattekyana 
ummoi mangayu-wot awad kad- 
ton ina nangayiiwan siat kapo- 
ngaton chit umillongan chat 

4. Ot mampokpok man sit 
awitonan kayu iningkau chad 
tumotoling-nga chingngoina ot 
manischung man chopya ining- 
kau chad mankakana taku. 

5. Nakitak si ogyatna ta sa- 
chit os-ossaan osya. Mangwa 
si silip ot ok-6yona chit chak- 
k6ya phyato-wot itopakna man 
chopya nipunta sit kawachan 
chat phusoi on mankakan. 



1. On a certain day there 
were men from Buaya going 
on a head-hunt to Patikian. 
As for those men come from 
Buaya they were thirty-five. 

2. When they came near to 
Patikian they sat down to rest 
a while and to eat in order to 
be strong to fight, as they said 
to themselves. 

3. However, there was a man 
from Patikian who had gone to 
cut wood and as for the place 
to which he had gone to cut 
the wood it was just above the 
resting place of the enemies. 

4. And while cutting his load 
of wood there were the sounds 
of talking people heard by him 
and when he looked down wist- 
fully there were people who 
were eating. 

5. Excessive was his fear for 
he was quite alone. He used a 
stratagem: he picked up a big 
stone and dropped it downhill 
hitting the place where the ene- 
mies were eating. 

1. nangayau, past of mangayau who goes headhunting. 
nampdipoy compare: Nampaipuam? Where do you come from? 

2. umachani one approaching, past ummachani. 

naphilog who or what is strong; a form with prefix ma does not exist. 
kancha means primarily "they say'' or "said" but probably also "they 

3. tai, also pronounced Ui or te. 

Ipattekyana, the final a is the connective particle. 
nangayu from kayu wood. 
kapongaton, for kapangaton? 


The Philippine Journal of Science 


6. Otchakona man chit phya- 
t6 pinachisananan nangkolis ; 
impakoinan : *'Mang6i kayon 
uchum lakud ta salewangam- 
min uchiim chaya!'' Ot nam- 
phubtik chat phusoi ta kancha 
nu achun takun manchogchog 
kan chicha. Awachoppai os- 
ossaan chit layakin nan-og-og- 
yat kan chicha. 

7. Ot awad kadchat6n phu- 
soi inaminchan tinainan chat 
kochonga alikamoncha ta sa- 
chit nakiyatanchan nakig-tot 
osya. Asiig-ka kanon cha na- 
amin niwalis. 

8. Ot mapotipot man chat 
phusoi nisukat chit layaki ot 
aminonan ubp5non chat ko- 
chonga tinainancha, ot mampa- 
lilid phophoyoi, on umoi ma- 
ngaya si phuyonan man-awit 
sichat uyos, kaman, topyai kan 
cha kayasag on tin6ponna. 

9. Sumaai phophoyoi chiyot 
iphyakanas chat susiinudna. 
Umolicha ot inaminchan inaya 
chat kochonga alikamon chat 
phusoi ot kod-kodwaoncha chat 

6. In dropping the stone he 
made at the same time a big 
noise shouting: "Some of you 
pass down river and we others 
get around them from up ri- 
ver!'' Then the enemies 
started to run thinking that 
many men were pursuing them. 
And there was left alone the 
man who had scared them. 

7. As for the enemies they 
left behind all their things for 
he had put them immediately 
into a fright. Their provision 
of boiled rice was all scattered. 

8. When the enemies were 
hardly gone he entered the 
place they had left and piled 
up all the things abandoned by 
them and then he returned to 
the village to fetch his compan- 
ions as carriers of those blan- 
kets, head-axes, spears, and 
shields which he had piled up. 

9. Arrived at home he told 
his relatives of the affair. They 
went back and took all the be- 
longings of the enemies and 
distributed their booty. 

4. tumotolingj compare tumling who talks noisily; past tinumling, 
chingngoif past of chongngoyon. 

5. ok-oyon what is lifted, past inok-oi. 

nipuntd, past of mijmntd (possibly from *maipuntd), compare Span. 

apuntar to aim at. 
kdwachan may have to be analyzed into ka-{-awad-\-an place of finding 


6. otchdkon what is dropped, past inotchak. 
impdkoi past of ipakoi what is shouted. 

salewanga/mmi from saleu-dngan-mi object of our surrounding, past 

19,2 Scheerer: Texts from Balbaldsang-Gindang 


10. Ot sumaai man chat phu- 
soi sit phophoyoicha maichaii- 
nai insaaicha tai natainan chat 
kochonga alikamoncha. Imph- 
yag-phyakachas chat kailian- 
chan : ''Nasuyagchan taktid 
Pattekyan ta ichakami inafof' 

11. Sia chiyot kochongchan 
iimoi mangaya-won ud Patte- 
kyan ta natilogcha. Achicha 
inum-ummoi ud Pattekyan 
ingkana si sumaai chat Meli- 
kanon umoi mampaamod ka- 
lingka. Sansaton maid mam- 
phuphusoi ta naamin ummamon 

10. And also the enemies 
arrived at their village bring- 
ing with them absolutely noth- 
ing for they had left behind 
their entire outfit. What they 
told to their villagers was: 
'The people of Patikian are 
brave for we were met by 
them/' they said. 

11. This, then, was the end 
of their going head-hunting 
to Patikian for they were 
scared. No more did they go 
to Patikian until the Ameri- 
cans arrived who came to tame 
the Kalinga. At the present 
time there are none warring 
with each other for they are all 
tamed people. 

7. tinainan, past of tainan what is left behind. 

kochong all without remainder, all to a finish; cf. Bont. kcechoeng chi 
this is all, Inib. ndkcheng finished. 

alikdmon equipment, personal belongings. 

nakiyatan has the sense of ''having done immediately." 

nakig-tot, past of makig-tot who frightens. 

asug boiled rice, with following connective a: asitg-ka; in rapid speak- 
ing asuka. 

8. mapotipot what has all but disappeared. 

nisukat, past of Tnisukat what takes the place of something else. 
aminonan probably amin all + on object of direct action + na his + n 

connective, the whole being adverbial to ubponon; past inamin; 

ubponon what is put on top of something else, past inuhpon. 
uyos cotton blanket, Inib. ulces. 

9. iiidya, past of dyon what is taken; Ayam nat lapis! Take that 

pencil ! 
kod-kodwdon object of distributing, past kinod-kodwd; probably from 
radical dud two: kad-kadud-on, the change from a to o not being 

10. maichaunai apparently from maid nothing + a connective -f unai ex- 

pressing a high degree, as "at all" in nothing at all. 
inafot, past of apton what is met. 

11. mangdya-won from mangayaii + on, 

180732 5 


The Philippine Journal of Science 




1. Osan laphi inummoi Se 
AXukana Ipinokpok namungwit 
se ikan ot opatchoet kennana. 
Ot anchimana koewoeen chi lap- 
hi manchukchuk-Aop kano-wot 
nasu-yop set ikid chit chentim. 

2. Anchin koewoeen man chi 
laphi ini-nau chit pfuoeye ot 
tumakchang chet pfuoeye ot 
chinukma-ana chet oApon AAu- 
kan ot illayugnat chet kaacha- 
Xoeman chet chentim ta pioeona 
no kanOna Se Ai^uken. 

3. Nomyoe Se A^uken naphi- 
log ot anchimana il-16k chet 
pfuoeya set liyang inawitna 
chet pfudeya ot intakchannat 
chit taAantag. 

4. Anchimana pillimaon chet 
pfuoeyan il-lok sichit chui^om 
kummapuyon Se AAuken ta 
achu chet phikoed set long-agna. 





1. On a certain night Si Alu- 
gan went to catch fish with a 
hook and four were caught by 
him. When it became mid- 
night he became sleepy, it is 
said, and he slept at the edge 
of the water. 

2. About midnight he was 
scented by the alligator and the 
alligator came out of the water 
going to the shore and caught 
hold of Alugan's leg and jump- 
ed with him into the deep of 
the water because it wanted to 
devour Si Alugan. 

3. But Si Alugan was strong 
and when the alligator was al- 
ready dragging him into a cave 
(under the water) he lifted it 
and went with it to the shore. 

4. When the alligator had al- 
ready sunk him five times into 
the deep Si Alugan became 
weak for many were the 
wounds on his body. 

AXukan, the form Alugan was said to be that given this name 

in Pinokpok, a town on the lower Saltan. 
Se, chit, pronounced also si, chet; I show the fluctuation by writing 
the vowel just as I heard it; ch in chit and other v^rords often 
sounds almost like ds; except before oe (as in choe) where it 
approaches ds. 
1. namungwit, compare Tag. bingwit, Bont. fengwit fishhook. 

kennan was stated to *be the past of kon-on object of catching, dif- 
ferent from kennan, past of kanon object of eating. 
2. ini-nau, past of inatvon object of smelling. 

tumakchang one going from the water to the shore; past tummakchang. 
chinukmd-an, past of chukmd-on object of seizing. 
illdyug, past of ildyug with what one jumps down. 


Scheerer: Texts from Balhaldsang-Gindang 


5. Anchimana adyan-dyanin 
matoi kenawotna chet attan 
chet pfuceya ot nilipsutan chet 
pfuoeya ta masikoeboen set 


6. Ot nilipstitana Se AAuken- 

not manayan set dudoAoina 
nomya achlna makaku9 ta ma- 
saxom chet phikoedna. 

7. Anchimana lumaos makat- 
lo ummoi chit pfuoey^ inlnap 
chet innoi AAuken ot inchasa- 
na chit nai-angon choe^^an AAu- 
kan. Ot siy9 inunu-tinod nan 
ininau chet innoina. Ot chi- 
natongna chet doAoi AAukan ot 
nansoyok set choya. 

8. Nomye anchimana lumik- 
na Se AAukan nanchit nasu-yop 
chingngoAna chet pfuoeye man- 
tatta-oA ta pioe5nan payau-on 
Se Ai^ukon. 

5. When he was almost dead 
he dug his fingers into the eyes 
of the alligator and the alli- 
gator let him free for its eyes 
caused it great pain. 

6. When Si Alugan was set 
free he went to his village but 
he could not walk for he had 
many wounds. 

7. When three nights had 
passed, the alligator went to 
search for the road taken by 
Alugan and found it by the 
dried blood of Alugan. And 
it followed the smell of his 
track. And when it reached 
the house of Alugan it hid it- 
self in the space under the 

8. But when Si Alugan 
awoke from his sleep he heard 
the alligator snorting for it 
wanted to cause him to come 

3. il-l6k, past inillok, what is caused to enter. 
inawit, past of awiton what is carried as a load. 

intdkchariy past of itdkchan what is taken from the water to the 

4. pillimaon, radical lima five; it appears herefrom that prefix pi (pin?) 

and suffix on are used to form with cardinal numbers derivatives 
indicating how often something is done. 
long-ag whole body; g with nasal resonance. 

5. adyen-dydni almost, Ilk. ngan-ngani. 

kenawoty past of kauton (from *kawoton?) what is put into. 

attd eye; note formation of genitive after noun ending in vowel: 
attdn chet pfuceyd eye of the alligator; compare genitive after 
noun ending in consonant: ikid chet chBnum edge of water. 

nilipsutan, past of lipsutan object of letting go. 

musikoebcen what is very painful. 

6. makaku9 one able to walk, say, do. 
masaKom what is numerous. 

7. makatloy the word ''night" is evidently understood. 
chceKa blood, Inib. chala. 

8. payau-on what is caused to come out, past said to be pinaya-u9. 


The Philippine Journal of Science 


9. Pfummangon Se AAukan- 
not ayana osan chakkoi^a duato- 
wot itup-tigna chet telid chet 
pf uoeya ot phinumtek set ikOu- 

10. Ot anchimana nianchat- 
chatong chet pf uyon AAuk8n set 
man-u-ukudchoe ta pioeonchoe 
no patayoxichoe chet pfuoey8. 

11. Ot manchat-chatongchoe 
man yc achuchoe ot kananchoen: 
**OA6gtak6n patayon ta achu 
tako.'' Ot inummoichce set 
ikouna ot kummuXubchoe se 
achoeAum-ma phito set ikid chet 
chantim set sog-6n chet ikouna. 

12. Maapfus chet phiton 
kingwoechoe ot pf unpf unanchoet 
set tupfun chet kayu. Maapfus 
chiyot ichoeyanchce chit osan 
aso-wot ipapanchcet chit phitu. 

13. Ot anchimana manta-Ui^ 
chet aso chingngo^^ chet pf uoeye 
ot manuchak chet pfuoeye ta 
ina kanon nomya naknat chit 
phltu. Ot pinatoichoeon chet 
pfuoeye. Nagangpoton. 

9. ayan object of taking, past indya. 

itup-ug object of throwing, past intup-ug. 
ikou (also ikau) dwelling-place. 
11. sog-ouy generally used as a preposition meaning ''near:" sog-6n chin 
doKoi near the house (if house is in view; otherwise sog-6n chit 
doKoi) . 

12. kingwce (from *kinwce?), past of kmi-on what is to be made; radical 

possibly kuci. 
ipdpan what is used as bait, past nipdpan; papdnan the place where 
a bait is used. 

13. naknd, past of maknd, what is caught (in a hole or on a hook). 

pinatoichoeon and nagangpoton: the final on in both expressions is 
probably equivalent to Iloko en, meaning "already," in such ex- 
pressions as nalpds aminen, all finished already. 

9. Si Alugan got up and took 
a big stone and hit the back of 
the alligator which ran away 
to its lair. 

10. And then the neighbors 
of Alugan came together to 
talk for they wanted to kill the 

11. And when they assembled 
they were many and they said : 
**We are sufficient to kill it for 
we are many.'' So they went 
to its lair and dug a deep hole 
at the edge of the water be- 
side its home. 

12. When the hole was fin- 
ished which they had made 
they covered it over with the 
leaves of trees. This done 
they went to get a dog and 
used it as bait in the hole. 

13. And when the dog barked 
it was heard by the alligator 
and the alligator ran to go 
and eat it but was caught in 
the hole. And then they killed 
the alligator. Finished. 

19, 2 

Scheerer: Texts from Balbaldsang-Gindang 





1. Se Lumawig na kawasana 
kaphilas maid kanona; hachoe- 
chit amana ya inana natoichoen 

2. Hachoechit kaama-an Lu- 
mawig kaina-ana yo kakaping- 
sana issanchoe Se Lumawig it- 
chen se kanona; ot chumakOA 
man Se Lumawig ummoi set 
tichum-ma dudoAoi ummOtang 
se pillak hichoechit uchum-ma 
taku ta tUi^tuXlian na chichoe. 
Haehit choekup chit inotangna 
nasui^ok-ka chuen gasot-ta pe- 

3. Ummoi man choechit taku 
imuson chit pillakchoen inOtang 
Lumawig; kananan: **Auni yan 
ta lyak man-ila se pillak se 
ma^^akptis kotnat pillak-ka inO- 
tangko ken chakayo/' Ot ma- 
doeyag manon kawoesa maid 
ochasana si pillak se mamdce- 
yadna choechit otangna. 

4. Osan laphi man ummoi 
Lumawig inakau chit pfuyon 
choe doeyug on napnO se apon- 
got. Ot haehit pottog chit 
ap5ngot ta inakau Lumawig 
kan ud doeyug tuAung gasOt-ta 
pesus. Ot haehit apongot im- 
doeyad nan amin sed choechit 
otangnat choechit tichum-ma 




1. Si Lumawig was absolutely 
destitute and had nothing to 
eat. Those parents of his 
were all dead. 

2. The uncle of Lumawig and 
his aunt and cousins would not 
give him any food and when he 
grew up he went to other vil- 
lages to borrow money from 
some people to whom he told 
lies. The total of his debt ex- 
ceeded two hundred pesos. 

3. When the people went to 
ask for the money which Luma- 
wig had borrowed he said: 
"Just wait, for I shall go and 
find the money to pay back 
that money which I borrowed 
from you.'' And for a long 
time after he absolutely could 
not find money to pay his debts. 

4. On a certain night Luma- 
wig went and stole the treas- 
ure-basket bf the Bayug family 
which was full of beads. And 
the value of the beads stolen 
by Lumawig from Bayug was 
three hundred pesos. And the 
beads he used as a means to 
pay all his debts to the other 

1, hachwchit, the same as later hichoechit, haehit, is sometimes pronounced 

with s for h {sachoechit, etc.); the' majority, however, is said to repu- 
diate this as incorrect; for oe compare notes to Salegseg phonetics in the 

2. issdn, expressing unwillingness, adverbial to itchsn; past inissan. 
itchdn, one to whom is given, past initchan; mangitod one who gives. 


The Philivpine Journal of Science 

5. Si Bayug went to the 
house of Lumawig in order to 
inquire about the place where 
Lumawig got the beads with 
which he paid the money he 
borrowed from the other peo- 

6. Then Bayug asked of Lu- 
mawig: ''From whom did you 
get the beads with which you 
paid the moneys you borrowed 
from the other people?'' said 

7. Lumawig replied: ''They 
are mine; if you like go and 
report it to the governor." And 
Bayug reported to the governor 
the beads stolen by Lumawig 
and the governor ordered the 
police to arrest the thief. 

8. Si Lumawig was brought 
by the police to be examined 
by the governor and he con- 
fessed to the governor (say- 
ing:) "I am the thief of the 
beads of the Bayug people." 
And the governor caused Si 
Lumawig to be conducted by 
the police to Bontoc and they 
put him into jail. 

3. imuson what is asked for, past inimus. 

kotnat probably contracted from ko my and sanat that. 

4. indkaUy past of akdwon object of stealing. 

chce doe/yug (Ilk. da Bayug) the Bayug family, the Bayugs. 
im dceydd, past of idcey9d, means of paying. 

5. nang-aR-an, past of mang-a/i-an place where, or person from whom, 

something is obtained; Si Pedro mang-aK-ak si tadceko Pedro is 
my purveyor for tobacco. 
namdcsyddnat, final t is an enclitic form of si (hi), likewise in fol- 
lowing inotangnat. 

6. sinnot from sinno si. 

namdce/i9dnot, the possessive suffix of the second person singular is 
no, not mo, 

7. uak from ud property + ^o mine. 

5. Se doeyug ummoi sichit 
do/ioi Lumawig ta ina chong- 
Aon chit nang-aA-an ud Luma- 
wig set apongot-ta namdceyed- 
nat chet pillak-ka inOtangnat 
choechit uchum-ma taku. 

6. Ot imuson man Joeyug 
kan Lumawig: ''Sinnot nang- 
a^^-am chit apCngot-ta nam doe- 
yadnot choechit pillak-ka inO- 
tangnot choechit uchum-ma ta- 
ku?" kanan man Joeyug. 

7. Ken-awat Lumawig: '*0t 
uak; no pioeom umoi ka icho- 
Aom kan kuphin-nach6L'' Ot 
ummoi doeyug inchoAom sichit 
kuphin-nachol chet ap5ngot-ta 
inakau Lumawig ot Se Kuphin- 
nachol ina impa-aAa chet nan- 
akau choechit p5lis. 

8. IsaaA man choechit polls 
Se Lumawig sinumai^an chet 
kuphin-nachol ot nampudnu Se 
Lumawig kan kuphin-nachol- 
la: "Sakon chit nangakau set 
apOngot choe doeyug." Ot im- 
pa-ichoe^^an kuphin-nachol Se 
Lumawig se choechit p5lis ud 
Pfuntok ot innigkcechoe chet 


Scheerer: Texts from Balbaldsang'Ginda7ig 


9. Ot maXakptis man osan 
chekun sinuma^^an ko6s Se Lu- 
mawig ot nakuoe chet lason 
Lumawig ot impaXQpfus koes. 
Ot imd^ken Lumawig kan 
koes-sa achi man-akau pal-yan 
ot ken-awat ko6s-sa : ' 'No man- 
akau-ka pal-yan ikanselme pal- 
yon seka ingkoenat matoi-ka/' 
Ot kai^san Se Lumawig set ili- 

10. Otin-kau man set Ilina 
nan-akau pai-y6n ot mangwce 
mat kapitan choet Ilemus ina 
inchai^om Se Lumawig ot pina- 
^Aa kuphin-naehol Se Lumawig 
hichoechit pOlis ot chumatong 
man Se Lumawig sinCma^^en 
chet kuphin-naehol ot masulit 
manon nampudno Se Lumawig 
ot kananan: **0n, nan-akawak'' 
ot painnigkan kuphin-naehol 
Se Lumawig set doKoi choechit 
suAchoecho ot imcEoekoen chet 
kuphin-naehol no chuan po- 
ison a^^aAkau ichceAanchoe ud 

9. When one year had passed 
he was examined by the judge 
and he gave explanations and 
was set free by the judge. 
And Lumawig told the judge 
that he would not steal again 
and the judge replied: "If you 
steal again we shall put you 
into prison until you die." And 
Si Lumawig went away to his 

10. And when he found him- 
self in his town he stole again 
and the capitan of the people 
of Lemus decided to go and 
report Lumawig and the gov- 
ernor caused the polioe to ar- 
rest him and when he arrived 
he was examined by the gover- 
nor and by and by Si Lumawig 
confessed and said: "Yes, I 
have stolen/' and the governor 
ordered him put into the house 
of the constabulary and the 
governor said that after twenty 
days he be conducted by them 
to Bontoc. 

8. sinumaHa, past of sumdKa, very probably from Span, sumdria first 

examination in court. 
impa-ichce/iany past of pa-ichceKan, cf . ichayan in notes to sect. 9 of 

The Eagle and the Child. 
Pfuntok, for a number of years after the advent of the American regime, 

the Kalinga towns whose speech is here recorded formed part of 

the present subprovince of Bontoc. 
innigkce, past of igkcB what is put into something. 
kansely from Span, cdrcel jail. 

9. chakun year, Iban. dagun; the above statement is evidently due to an 

error; the rights of an accused person are well established under 

American law all over the Islands. 
koeSy from Span, juez, judge. 
nakuce, past of makuoe one saying, doing. 
lasoriy from Span, razon reason. 
imdoekoe, past of idcekce what is told; cf. iphyakd in Bal. 


The Philippine Journal of Science 


11. Ot osan laphi man kanan 
Lumawig-koe: *lyak umispfo'' 
ot in-kauchoe man set lasin ina- 
yan Lumawig chet Joechoeng- 
ngot songpatonat taki<:ai chet 
suAchoecho. Ot mangwoe mat 
suAchoecho inchongpa^nat doe- 
yoneta set pfuang Lumawig ot 

12. Ot mangwoe man choechit 
pfuyon Lumawig ichoe innaya 
lachoegna ot inchoeAanchoet chit 
ihchoe ot pa-isan choet Luma- 
wig se chuan pfuoX-Aok. Lum- 
maus mat chuon aA-a^kau in- 
nildon choet Lumawig. 


1. Chat man-aama tuyu na- 
pangugcha. Inaikau chi aikau 
incha man-iwa-ai sit payau. 

2. Sit osan ai-aikau ummoi- 
cha nan-iwd-ai sit payau-wot 
ingkaucha pon sit payau um- 
muchan-not man-iwa-aicha. 

3. Sit amacha man-iwa-ai sit 
tupping, sit phuphya-i man- 
iwa-ai sit sag-on-na, sit laya-i 
man-iwa-ai sit phi-ik. 

11. And one night Lumawig 
said : ''I go to make water/' and 
when they were outside Luma- 
wig took a bolo (cleaver) and 
cut into the arm of the soldier. 
And the soldier responded by 
sticking his bayonet into the 
stomach of Lumawig who died. 

12. And the town-mates of 
Lumawig decided to go and get 
his body and to transport it to 
his town and they killed two 
pigs at his funeral. When two 
days had passed they buried 



1. The father and his two 
children were diligent workers. 
Day after day they went to 
work in the rice-field. 

2. On a certain day they had 
gone to work in the rice-field 
and finding themselves in the 
field the rain fell, yet they 
worked on. 

3. The father was working 
near the stone-wall, the daugh- 
ter (lit. woman) worked by 
his side, and the son (lit. man) 
on the other side (of the field) . 




paydn again, Bal. would use here umphos, with same meaning. 
mangwce, cf. mangwd, in notes to sect. 11 of The Eagle and the Child. 
Ilemus the people of the town of Lemus (Limus) on the Saltan east of Saleg- 

suKchcecho from Span, soldado, here applied to members of the constab- 

songpaton what is cut, past sinungpat. 

inchongpa/iy past of ichongpa^ instrument of sticking into. 

pa-lsan who is made the recipient of the funeral honors consisting 
in the slaughtering of two pigs; the souls of the animals are supposed to 
accompany the deceased who is thus appeased and kept from making 
trouble for his bereaved friends. 


Scheerer: Texts from Balbaldsang-Giyidang 


4. Sat man-iwa-ayancha in- 
ingkau ngas-ngas-sa nataknang 
ot mangwa mat uchan pina- 
chagson-nat pita ot magchai 
chit ngas-ngas ot matab-unan- 

5. Ot sit laya-i chingngoi chit 
nagchai-yot mantuwili pon si- 
chat amana an suniidna naid- 
cha pon ot ummoi pon umacha- 
ni nagchayancha ot man-iphil 
ot chongyon man chat uchum- 
ma taku ini-moscha an siya. 

6. Ot iphyakanan nagchayan 
chat amana an suniidna ot sit 
takun na-ngimiis nampaoi ot 
umali chat taku sit ili. Ot 
chummatong man chat taku an- 
nan maniiyung chit laya-i. 

7. Ot phya-ungancha ot ayan- 
cha chicha. Ot ayancha man 
chicha natakucha. Ot phuli- 
kancha phuphya-i ya amana sit 


4. At the place where (the 
former two) were working 
there was a steep slope which 
was very high, and as the rain 
had made the soil heavy, the 
slope collapsed and buried 

5. When the son heard the 
slide he turned to look for his 
father and sister who were no 
more there and he went to ap- 
proach those who were buried 
and he cried and he was heard 
by the other people who asked 

6. And he told them that his 
father and sister were buried 
and the people who had asked 
shouted and the people from the 
town came. And when the peo- 
ple arrived there was the son 

7. And they dug for them and 
got them. And when they got 
them they were alive. And 
they carried the woman and 
her father to the town. 

man-dma a term used to designate a group consisting of a father 
and his child; if there are two or more children man-adma. is said; note 
added tuyu, three, indicating the number of persons comprehended 
in the group. (Sect. 1) 
inagchaydUy past of Tuagchaydn who is caught in a landslide. 

1. incha went-they, past of icha go-they. 

inaikau, the second i is the sound represented by Ji in Sal. 

2. ummuchan, past of umuchan it rains. 

3. tupping the stone- wall supporting the next higher rice-terrace. 
sag-on, cf. sog-6n in note to sect. 11 of story of Si Alugan. 

4. pinachagson, past of pachagsonon what is becoming heavy. 
pita earth, soil. 

magchai what collapses, past nagchdi. 

5. naidcha pon: naidcha not-there-they, pon seems to be added to make 

the statement more impressive after the manner of Ilk. met. 
an siya to him, compare kan sia in sect. 3 of Changatag and the Monkeys. 


The Philippine Journal of Science 

8. Ot chat uchum-ma ining- 
kau sit ili ancha gai-ya: 
'Tatoi!'' Ot naammin chat 
taku ummoi sit nagchayan. 

9. Ot imusoncha pon: 'Tho- 
ompon si patoi?'' — yot ipa-au- 
ascha gai ilan chit nagchayan. 
Ot sit amacha na-aliwong-ngot 
man-u-iikudj chat apingsan-na 
ot patoyonchat opat-ta phoyok 
ta isichan chat takun nanchat- 

10. Ot nagamputcha pon nan- 
gan aisancha. Ot naanai liman 
phuyan pon naid chit mansikab 
sichat nagchayan. 

11. Ot ilapocha paiyan-nat 
inaikau chi aikau man-iwa-ay 
tai pia-onchat chit magamput 
chit payau ta asiyot matoi chit 

12. Ot chuan phuyan chit is- 
sapon man-illong chit ama. 
Na-aliwong si opat-ta phuyan 
ot matoi. Ot man-u-iikudchat 
a-pingsan-na ot patoyonchat 
opat-ta luang ta isichanchat 
taku-wot chuan ai-aikau pon 
inilphoncha ot mangwacha si 
phoyoi sit lap-at chit lophon. 

8. And the others who had 
stayed in the town just said: 
"Killed!", and all the people 
went to (see) those who had 
been buried by the landslide. 

9. And they asked: "Not 
killed?'' and passed on just to 
have a look at those buried by 
the landslide. And the father 
was sick and his cousins talked 
together and they killed four 
pigs for food of the people who 
had arrived. 

10. And when they had fin- 
ished eating they went away. 
And it took five months to cure 
the pains of those buried under 
the landslide. 

11. And they again began to 
work day by day for they 
wanted to finish the rice-field 
before the old man died. 

12. And for two months the 
father did not take a rest. He 
was in bad health for four 
months and then he died. And 
his cousins talked together and 
killed four water-buffaloes as 
food for the people and after 
two days they buried him and 
made a house over his grave. 

6. sunud brother or sister. 

anna, probably a deictic expression similar in force to French voild. 

7. phulikan, past phinulikan, compare Ilk. huligan what is carried between 


8. dncha gai said-they just, cf. Ilk. konada laeng; Gin. dncha = Bal. 

kdncha. ya after gai is connective particle a plus consonantal initial 
drawn over from gai. 

9. isichan any boiled food taken with rice; preferentially meat. 

10. nagamput or nagangput finished; pronounce na-gamput. 
naanai was said to be the past of umdnai what is sufficient. 

11. pia-on object of liking, past pinia. 

mayong-dg old man; the y of this word is the sound represented in 
Sal. by A. 

i9»2 Scheerer: Texts from Balbaldsang-Gindang 203 

13. Ot sit asawan chit natoi 13. And the wife of the dead 

issapon lumawa si simphuyan. man did not go out for one 
Ot sit payau chit natoi nampa- month. And the rice-field of 
yau chit anakna. the deceased became the rice- 

field of his children. 



1. itnoeg ta kasigiidan na ta-o 1. The Itneg are the original 
di Abra. Dad k5dil dangi kiu- people of Abra. Their skin is 
ti kan dad biiok dangi ataddo brown and their hair is long 
kan nangisit. and black. 

2. Daddi babai-ye agikkuada 2. The women put on beads 
ta batek aglikmot kandi ulo, around their heads, hands, and 
ima kan daddi tengnged dangi ; necks, and the men put a tur- 
ket daddi lalaki-e agikkuada ta ban around their heads, and a 
ayabong aglikmot kandi ulo girdle and gee-string around 
dangi, kan barikoes kan baal their loins, and they suspend 
aglikmot kandi sikoet dangi kan the short bolo from their belts, 
agsaklitda ta immoko kandi ba- 
rikoes dangi. 

3. Daddi lugak dangi na- 3. Their clothes are made 
appia nga kapas ket abuenda of cotton and they weave also 
payen daddi lugakda nen. their clothes themselves. Their 
Daddi ayabung dangi ma-appia turbans are made of bark of 
nga ukes ta malit-i kayo. the balete tree. 

12. issapon man-illong did not rest; it appears that while Gin. uses issd 
as a negative particle in connection with verbal forms {issdak pon 
man-illong 1 do not rest, issacha pon lumawa they do not go out), 
Bal. uses instead achi (almost adzi) : umachik, umaehika, uma^ 
china umillong I do not, thou doest not, he does not rest ; in Sal. texts 
we find: achina makakua he could not walk, and issdnch(» itchdn would 
not feed. 
plioyoi sit lap-dt chit lophon, this probably ritualistic structure is known 
by the special term bagong and was said to be destined to protect 
the grave against rain. 

1. kasigiidan origin, beginning, birth, thus kasigudan na td-o may also 

mean "native people." 

2. agikkuada put-they, compare Ilk. ikud what is put in some place. 
badl gee-string, Ilk. hoAg. 

sikoet loins; the mixed vowel in this word, as in Hnceg, barikoes, is 
short and similar to the one in French neuf; when long, it resembles 
eu in French peur. 


The Philippine Journal of Science 

4. Agyanda ta abong nga na- 
appia nga kayo, kawayan, kan 
golon. Unneg nid abong dangi 
oad ta kal-ang nga na-appia nga 
lota. Ket sidioe nga kal-ang 
uad ta ta-lo a lilona nga nagsa- 
sango koet dublena ta agtuga- 
wana nidi banga-ngoe. 

5. Ket no bolan ta ukop, 
kiang kan ladau aga-apoida 
kandi dud-duagan dangi koet 
aga-anidoda oedoeng ta adi lum- 
tau di init-te. Ket agtotogau- 
da kandi likmot nidi apoi-ye. 

6. Koet no madatngan ta bo- 
lan ta manaba mangrugida nga 
aguangoel kandi kab-kabasaan 
danin. Koet daddi lalaki-ye 
daida ta agarado kan agpailid ; 
koet daddi babai-ye daida ta 
agoto ta kanenda koet mapanda 
ibaonanda dad asawa dangi no 
mamatoon. Koet no malpas ta 
panagroeoep ta pagei daddi ba- 
bai-ye agaboda ta lugak koen 
owoes. Koet daddi moet lalaki- 
ye mapanda agayoan ta nuang, 
baka, kabayo, kan lyas. 

4. Their dwellings are houses 
built of wood, bamboo, and 
cogon grass. Inside of their 
houses there is a fire-pot made 
of clay. And that fire-pot has 
three heads which bend toward 
each other and this is the sit- 
ting-place of the cooking-pot. 

5. And during the months of 
December, January, and Feb- 
ruary they will make fire in 
front of the houses and warm 
themselves while the sun does 
not rise. And they are sitting 
round the fire. 

6. And when there has ar- 
rived the month of May they 
begin to work in their irrigated 
rice-fields. And the men they 
plow and harrow, and the wo- 
men they cook their food and 
they go to provide their hus- 
bands with food at noon-time. 
And when the planting of the 
rice is finished, the women 
weave clothes and blankets. 
And also the men go to take 
care of the buflFaloes, cows, 
horses, and pigs. 

unneg nid, also unneg nidi is said. 

kal-ang, Tag. kaldn the typical Philippine utensil, consisting in a heavy, 
\vide-oj)en, flat fire-pot of clay on the flat floor of which the fire is 
kindled and on whose upper rim three projecting points are bent 
inward so as to serve as supports for the superimposed cooking-pot. 

dud-duagan, cf. Ilk. duag roof over entrance to house. 


Scheerer: Texts from Balbaldsang-Gindang 


7. Koet no madatngan ta pa- 
naglaani, nga dublena ta bolan 
ta ukop, agkakalsoda, daddi ba- 
bai-ye nga mapanda aglaani; 
koet daddi meet lalaki-ye daida 
ta agbetek kan agbonag-ga ag- 
yodong kadad pagei-ye. 

8. Koet no malpas ta laani 
ikuada daddi pagei dangi kandi 
baang dangi ta sannon mala- 
ngo. Koet no malango di pagei 
nin, ipasakaida kandi alang da- 

9. Koet kalpasan datoe nga 
maappia rugiandan ta agappia 
kadadi an-annongda nga anna- 
wid. Koet no oad ta masakit 
kadaida agbalawada oenno ag- 
pal-laanda ta sannon malai- 
ngan di masakit-te oenno paka- 
wanena nidi nangkaro dobli 

10. Ket dato Itnoeg-e nag- 
gagitda moet nga aguangoel. 
Nalaengda meet nga agibaon 
kadaddi anak dangi nga agadal. 

11. Adu pai lameng ta anna- 
wid ta Itnoeg. Noem adiak ml- 
baga nga lisan ita. 

7. And when comes the time 
for harvesting the rice, which 
is the month of December, 
they go to the fields, the wo- 
men in order to cut rice, and 
the men, on their part, they 
bundle and go to and fro carry- 
ing home that rice. 

8. And when the cutting of 
rice is finished they put the 
rice in stacks in order to dry. 
And if the rice is dry they lift 
it up into their granaries. 

9. And after this work they 
begin to make their ceremonies 
which are customary. And if 
there is a sick person among 
them, they have the ceremony 
called ''balawa'' or the lesser 
ceremony called "pal-laan,'' so 
that thus may get well the sick 
person, or that he may be freed 
from his malady (by the grace 
of him who caused it through 
his evil influence). 

10. And the Itneg are also 
industrious as workers. They 
are likewise intelligent enough 
to send their children to study. 

11. Many more yet are the 
habitual occupations of the It- 
neg. But I cannot relate them 
all now. 


The Philippine Journal of Science 

The personal and possessive pronouns in Tinggian, Balhaldsang, Salegseg^ 

and Gindang. 


Singular. . 

First person. -_ 
Second person . 
Third person .. 


diaken, -ak 
dika, -ka 


sakon, -ak 
seka, -ka 



siak, -ak j sa-on, -ak 
seka, -ka j 8i-a, -a 


First person inclusive ; ditai, -tayo 

First person exclusive.-- I dikame, -kame 

First person dual dita, -ta 

Second person dikayo, -kayo 

Third person daida, -da 

chitako, -tako 
chakame, -kame 
chitd, -ta 
chakayo, -kayo 
chicha, -cha 


First person. -_ 
Second person . 
Third person . . 

ko, -k 
mo, -m 

ko, -k 
no, -m 


First person inclusive . 
First person exclusive . 

First person dual 

Second person 

Third person 











, chita-o, -ta-o 
chi-ame, -ame 
: (si-a sa-on) -ta 
i chi-ayu, -ayu 
I chicha, -cha 

o, -k 
no, -m 








Fig. 1. Map of the subprovince of Kalinga. 




Of Dorking, England 


Genus SAMIA Hiibner 
Samia cynthia Drury. 

Plate 1, fig. 1, larva, third ? stage; fig. 2, larva, fourth ? stage; figs. 
3 and 4, larva, fifth ? stage, dorsal and lateral aspects; fig. 5, 
larva, sixth ? stage; fig. 5a, food plant. Larva of Samia pryeri, 

Japanese names: Sh/inju-san; aya-nishiki; mikazuki, 

Phals^na cynthia Drury, 111. Exot. Ent. 2 (1773) pi. 6, fig. 2; Kirby, 
Cat. Lep. Het. (1892) 748; Leech, Proc. Zool. Soc. London (1888) 
634, No. 255; Trans. Ent. Soc. London (1898) 264, No. 4; Hampson, 
Moths Brit. India 1 (1892) 16; MooRE, Lep. East. Ind. Co. (1859) 
pi. 20, figs. 3, 3a; Nawa, Insect World (Konchu Sekai) 10 (1906) 
366, pi. 9, larva, pupa, imago cf; Matsumura, Cat. Insect. Japan 
(1905) 47, No. 383; Thousand Insects of Japan [Nihon Senchu 
Dzukai (Jap.)] (1909) suppl. 1, 40, No. 65, pi. 6, fig. 1, d"; Jordan, 
Seitz's Macrolep. Faun. Pal. 2 (1911) 212. 

Attacus walkeri Felder, Wien. Ent. Mon. 6 (1862) 34; Jordan, 
Seitz's Macrolep. Faun. Pal. 2 (1911) 212. 

Attacus pryeri Butler, Proc. Zool. Soc. London (1878) 388; 111. Typ. 
Lep. Het. 3 (1879) pi. 43, fig. 5; Pryer, Trans. Asiat. Soc. Japan 12 
(1883) 53, No. 196; Sasaki, Insects Injurious to Japanese Trees 
[Nihon Jumoku Gaichuhen (Jap.)] ed. 3 (1910) pt. 3, 74, pi. 198, 
larva, cocoon, imago ?; Jordan, Seitz's Macrolep. Faun. Pal. 2 
(1911) 213, pi. 33, fig. a, d". 

Jordan ^ describes the larvae and pupae of the genus Samia as 
follows : 

Larvae powdered with white; with six rows of fleshy thorns, which bear 
sparse bristles, the dorsal processes longer than the lateral ones in the 
earlier stages, later on the difference smaller. Cocoon long, pointed at 
both ends, usually wrapped up in a large or several small leaves. Pupa 
without bristles at the anal end. 

* The editors have been unable to verify the quotations in this article. 
'Seitz's Macrolep. Faun. Pal. 2 (1911) 212. 

180732 6 209 

210 '^h^ Philippine Journal of Science 1921 

Two species, one of which is confined to India while the other occurs 
from Japan to India and eastward to the Sula Islands,, and is also ac- 
climatised in several localities in Europe and North America. 

Jordan describes the larva of Samia cynthia Drury as follows : 

Larva at first yellowish, then white or greenish or bluish green, with 
black dots; the dorsal processes in the later stages about as long as the 
distance between two dorsal processes of the same segment; the processes 
of the different segments almost equal, bluish. Polyphagous, chief food- 
plants: Ailanthus, Ilex, etc.; prefers Syringa in Europe, but also takes 
Prunus, Pirus, Laburnum, and other plants. The cocoon consists of an 
outer layer of loose silk and an inner dense web. The silk, from which a 
very tough cloth i^ made, cannot be reeled, but is carded and then spun. 
It is coarse and not valuable for export. True cynthia Drury is confined 
to the Malayan districts. On talearctic territory two forms are found: 
pryeri Butl. (33a) inhabits Japan. The proximal band of the forewing 
more or less convex posteriorly, without long teeth on the median veins, 
the discal band twice deeply incurved, only faintly reddish outside the white 
line, at least the red line is never sharply defined; on the hindwing the 
discal band is deeply incurved below the half moon. The cocoon is grey 
or yellowish white. The larvae (tscho-san) are found especially on Ilex 
rotunda,^ Ailanthus glandulosa and Phellodendron amurense.^ The speci- 
mens from North and Central China, walkeri Fldr.^ are distinguished from 
pryeri by the discal band of both wings having a sharper outer edge and 
by the long median teeth of the proximal band of the forewing. This 
form goes northward to Manchuria and Corea, and is much kept domesti- 
cated for the sake of its silk, e. g. in the provinces of Shantung and Che- 
kiang. This form was introduced into Europe in 1845 and has been domes- 
ticated in France with some success. In the Northern districts the species 
has only one brood, in the Southern countries several broods in a year. 
The second species of the genus, S. lunula Walk., is likewise easy to breed, 
and crosses between it and S. cynthia are also known. 

Jordan gives the name tscho-san to the larva of pryeri. I 
have never heard this name. *'Tscho" is certainly not Japanese 
but may be Chinese and probably refers to walkeri Felder, which 
occurs in China. The imago is known under the Japanese 
names of shinju-san (which name is probably given because the 
larva feeds on shinju), aya-nishiki, and mikazuki, '*new moon/' 
referring to the crescent on the wings. 

The larva of Samia pryeri Butler is figured in what I believe 
to be the third, fourth, fifth, and sixth stages (Plate 1, figs.l 
to 5) . A brood of young larvae was taken about September 15, 
1900, at Kobe, Settsu Province, Honshu, on kitsagi (Cleroden- 
dron tricotomum Thunb.), one of which was selected to be 

"^ Ilex rotunda, named in Japanese fukura-shiba and kurogane-mochL 
* Phellodendron amurense, named in Japanese kiwada. 

19,2 Wileman: Japanese Lepidoptera, VI 211 

figured in its various stages. This larva was figured in the 
third ? stage (Plate 1, fig.l)) on September 19; in the fourth ? 
stage (Plate 1, fig. 2), on October 1; in the fifth ? stage (Plate 
1, fig. 3, lateral aspect; fig. 4, dorsal aspect), on October 9; in 
the sixth ? stage, when it measured 66 millimeters in length, 
on October 26, 1900 (Plate 1, fig. 5). This larva and the rest 
of the brood were nearly all full-grown by October 26, but they 
all failed to develop imagoes after pupation as they were in- 
fested by the larvse of an ichneumon fly. Several specimens 
of this fly emerged from the cocoons of Samia pryeri in April, 
1901, and another emerged on June 27, 1901.^ 

A male imago was bred from a larva of Samia pryeri taken 
at Yoshino, Yamato Province, Honshu. This larva pupated in 
October, 1900, and the imago emerged on June 15, 1901. I also 
bred a specimen at Kobe, Settsu Province, Honshu, in August, 
1901. Other food plants of the larva are maple, momiji {Acer 
palmatum Thunb.), and plantain, basho (Musa basjoo Sieb.). 

Nawa gives the following notes on Attacus cynthia Drury: 

The larva ° feeds upon shinju [Ailanthus giandulosa Desf.] ; nurude 
[Rhus semialata Murr. var. osbeckii DC] ; konzui [Latin name unknown, 
not given by Matsumura in his Shokubutsu Mei-i]. The body, head and 
ventrum are pale yellow; pale indigo, fleshy tubercles on each segment 
which are clothed with a white flourlike substance. 

The larva and the imago appear twice in the year. The first brood 
of the imago emerges about June or July, the second brood about 
September, or October. The larvae which emerge from the second brood, 
when full grown, spin their cocoons and hibernate in the pupal stage, 
the imago emerging at the commencement of the following summA, when 
the ova are then deposited. 

I find that in the final stage of the larva the white powder 
on the tubercles disappears. It seems to appear after the third 
? stage when the larva turns from yellow to white. 

Sasaki ^ gives descriptions and figures of the larva, cocoon, 
and imago of Attacus pryeri Butler (Attacits cijnthia Drury) 
and a short recoi^d of its life history. He says: 

The larvae emerge from the ova about middle or end of June, and 
are full-grown by the end of July. The imago emerges about the com- 
mencement of August and dep^osits ova which hatch in about three weeks at 

* This ichneumon fly has been identified by Claude Morley, Entomologist 
43 (1910) 11, as Pimpla luctuosa Smith, Trans. Ent. Soc. London (1874) 

° The larva figured by Nawa measures 72 millimeters. 

^Insects Injurious to Japanese Trees [Nihon Jumoku Gaichuhen (Jap.)] 
ed. 3 (1910) pt. 3, 74, pi. 198. 

212 The Philippine Journal of Science 1921 

the end of September when the seconCl brood of larvae appear. These 
larvae are full grown by the end of October when they pupate and 
hibernate in the pupal stage, their imagoes emerging at the commencement 
of June of the following year. The ova deposited by these imagoes gen- 
erally hatch in eight or nine days. 

Pupa, — The pupa is contained in an elongated, brown cocoon, 
broad at the bottom and tapering above, which is made of 
extremely tough silk. It is suspended by a long, tough silken 
fiber to the bough of a tree, somewhat in the same manner as 
the cocoon of Rhodinia fugax Butler. 

Local distribution, — Honshu, Musashi Province, Yokohama, 
July (Pryer) : Musashi Province, Tokyo, July and August 
(Wileman). Kyushu (recorded by Leech, and my collector has 
gathered cocoons there). Matsumura records the species from 
Honshu, Shikoku, and Kyushu. 

Time of appearance. — Larva, June to July, September to 
October; imago, June to October. 

General distribution. — Samia cynthia, Malayan districts; S. 
pryeri, Japan; S. walkeri, northern and central China, Man- 
churia, Korea. (Jordan.) 

Genus RHODINIA Staudinger 

Rhodia Moore, Proc. Zool. Soc. London (1872) 578 (nom. praeocc). 
Rhodinia Staudinger, Rom. Mem. Lep. 6 (1892) 327. 

Rhodinia fugax Butler. 

Plate 1, fig. 6, larva, fifth ? stage; fig. 7, larva, sixth ? stage; fig. 8, 
paired dorsal tubercles on segment 4; fig. 9, tubercles on segments 
ll and 12. 

Japanese names: Usutahi-ga, yama-bishaku, yama-kamasu, ajima-ni- 
shiki, yama-biku. 

Rhodia fugax Butler, Ann. & Mag. Nat. Hist. IV 20 (1877) 480; 
111. Typ. Lep. Het. 2 (1878) 17, pi. 26, fig. 1, ^; Pryer, Trans. 
Asiat. Soc. Japan 12 (1883) 52, No. 193; Leech, Proc. Zool. 
Soc. London (1888) 633, No. 251; Staudinger, Rom. Mem. Lep. 6 
(1892) 327; Leech, Trans. Ent. Soc. London (1898) 268, No. 13; 
Staudinger and Rebel, Cat. Lep. Pal. 1 (1901) 126, No. 1027; 
MATSUMURA, Cat. Insect. Jap. 1 (1905) 47, No. 386; Nawa, Insect 
World [Konchu Sekai (Jap.)] 10 (1906) 277, pi. 8, larva, pupa, 
imago ?; Matsumura, Thousand Insects of Japan [Nihon Senchu 
Dzukai (Jap.)] (1909) suppl. 1, 41, No. 66, fig. 2, cf; Sasaki, 
Insects Injurious to Japanese Trees [Nihon Jumoku Gaichiihen 
(Jap.)] ed. 3 (1910) pt. 2, 43, pi. 95, ova, larva, pupa, cocoon, imago 
c^; Jordan, Seitz's Macrolep. Faun. Pal. 2 (1911) 213, pi. 34, c, 
d*, 2; Wileman, Trans. Ent. Soc. London (1911) 344, No. 327. 

Saturnia diana Oberthur, Bull. Soc. Ent. France VI 6 (1886) 47; 
Staudinger and Rebel, Cat. Lep. Pal. 1 (1901) 126, No. 1027a; 
Jordan, Seitz's Macrolep. Faun. Pal. 2 (1911) 213, pi. 34, c, d*, 2. 

19,2 Wileman: Japanese Lepidoptera, VI 213 

The larva figured (Plate 1, figs. 6 and 7) was taken in May, 
1901, at Hiyeizan, near Kyoto, Yamashiro Province, Honshu, 
on dwarf oak {Quercus serrata Thunb.), Japanese name kunugi. 
It was figured on May 29 after its fourth ? molt (Plate 1, fig. 6) 
and again on June 23 after its fifth ? molt (Plate 1, fig. 7). 
It pupated on June 27, and a male imago emerged, which I 
identified at the South Kensington Museum as a dark form of 
Rhodinia fugax Butler. 

Nawa ^ records the life history of R. fugax Butler and gives 
a figure of the I^rva after its sixth molt, of its pendant cocoon, 
and of the female imago. 

Sasaki « also gives descriptions and figures of the ova, larva, 
pupa, cocoon, and male imago. He says that — 

* * * the larvae emerge from their ova at the end of April and 
feed upon dwarf oak, nara [Quercus glandulifera BL] and dwarf oak 
kunugi [Quercus serrata Thunb.]. They are full-grown by the end of 
June and the imagoes emerge at the end of October or commencement of 
November when the female oviposits. 

Jordan ^ describes the larva, cocoon, and pupa of the genus 
Rhodinia as follows : 

Larva almost naked, granulose, the six warts of the prothorax separated, 
but the two dorsal ones close together, the dorsal warts of the metathorax 
the largest, segment 11 with one dorsal wart instead of 2. Cocoon egg- 
shaped, but truncate at the upper end, and attached to a leaf or slender 
twig on one side so that the cocoon appears stalked at one corner (pitcher- 
like) , dense, without outer loose silk. Pupa attached by the hooked bristles, 
which are placed close together, to a loosely woven transverse wall, which 
stands close to the apex of the cocoon. The full-grown larva as well as the 
pupa makes a loud chirping noise when disturbed. Distributed from the 
Himalayas to Amurland and Japan. 

The following description given by Nawa of the larva of 
Rhodinia fugax in its various stages has been translated by me 
from the Japanese text: 

On emergence from the ovum. — The head and segment 1 black; segment 
2 and all the others light yellow; a black mediodorsal line; body clothed 
with black hairs. 

Three or four hours after emergence from ovum, — The dorsum becomes 
almost entirely black; laterally yellow. 

After first molt. — The larva becomes yellow below the subdorsal line; 
black-haired tubercles appear on the body. 

'Insect World (Konchu Sekai) 10 (1906) 277, pi. 8. 
* Insects Injurious to Japanese Trees [Nihon Jumoku Gaichuhen (Jap.)] 
ed. 3 (1910) pt. 2, 43, pi. 95. 

'Seitz's Macrolep. 2 (1911) 213. 

214 The Philippine Journal of Science 1921 

After second molt, — The black dorsal area becomes yellow, and the lat- 
eral yellow becomes black; the tubercles which have greatly increased in 
length are arranged as follows: Segments 1 and 12, four tubercles each; 
segments 2 and 11, six tubercles each; segment 3, two dorsal tubercles; 
segment 11, one dorsal tubercle; below the spiracles on segments 2 and 11 
the tubercles are blue, each emitting four or five black spines. 

After third molt. — The whole body becomes pale yellowish green and 
occasionally there is a black lateral stripe. 

After fourth molt. — There is no change whatever. 

After fifth molt. — The larva loses all the long spined tubercles, with the 
exception of the two dorsal tubercles on segment 3 and the one dorsal tuber- 
cle on segment 11. At this stage the larva, when alarmed at any thing, 
makes a noise something like the sound of "chii." 

After sixth molt. — When full grown the larva is greenish yellow; 
ventrum bluish green; a subspiracular stripe above which there is on each 
segment one small blue tubercle; two longer dorsal tubercles on segment 
3, the mediodorsal shorter tubercle on segment 11 still remains; the poste- 
rior margin of all segments paler in color; on the dorsum and body there 
are minute warts. When full grown it spins a green cocoon the upper 
end of which has a hole which seems as if it had been cut out in a circle; 
the cocoon is suspended by a thread attached to its upper angle and hangs 
down from the branch. The larva emerges from the ovum in April and 
feeds on nara^ [dwarf oak {Quercus glandwlifera Blume) ] ; kunugi, [dwarf 
oak (Quercus serrata Thunb.)]; kuri, [chestnut (Castanea vulgaris Lam. 
var. japonica DC.)]; kashi [Quercus acuta Thunb.]. 

It pupates at the end of June or beginning of July, and the imago 
emerges in October or November. The female deposits its eggs on the 
twigs of the tree or on the cocoon and they hatch in April as previously 
stated. The imago is single-brooded. 

The larva as figured by Nawa measures about 87 millimeters. 
According to IVEatsumura it also feeds on keyaki (Zelkowa acu- 
minata Lindl.) and kashiwa (Quercus dentata Thunb.). Pryer 
gives cherry and other trees. Jordan gives Phellodendron 
w^hich, if it be the same as the Japanese species (Phellodendron 
amurense Rupr.), is called kiwada. 

The following description of the larva is taken from my 
original figures: 

After fourth ? molt— Plate 1, fig. 6; figured, May 29. 
Length, 54 millimeters. Head light green; body apple green; 
ventrum, all legs, prolegs, and claspers darker green, speckled 
with minute yellowish spots; spiracles brownish; a yellowish 
white, subspiracular line; paired, spinous, dorsal tubercles on 
segments 2 to 13, those on segment 2 (next to the head) being 
blue; the tubercles on segments 4 and 12 are longer and more 
prominent than the rest; a spined, midlateral, blue tubercle on 
segment 2; a series of subspiracular, spined, blue tubercles on 

19, 2 Wileman: Japanese Lepidoptera, VI 215 

each segment from 2 to 12; two, spined, blue tubercles on the 
anal segment 13, above the subspiracular series; one spined, 
blue tubercle above each thoracic leg, below the subspiracular 
series ; all the spined tubercles of segment 2, those of the anus, 
those of the subspiracular series, and those above the thoracic 
legs are blue; all the rest are green. 

After fifth molt. — Plate 1, fig. 7, figured, June 13. Length, 
64 millimeters. Segment 2 edged with yellow; body and head 
lighter in color, inclining to greenish yellow ; ventrum as before ; 
subspiracular line not so prominent; all the spined tubercles 
have disappeared with the following exceptions: There still 
remain long, spineless, paired tubercles on segment 4 and a 
single, shorter, spineless tubercle on segment 12; also the sub- 
spiracular series is represented by a spineless blue wart on each 
segment from 2 to 12 and by similar spineless, blue warts above 
each of the three thoracic legs. The dorsal and anal series of 
spined tubercles, which have disappeared, are represented by 
small yellow warts. My original figure of the larva agrees with 
those given by Nawa and Sasaki. 

Cocoon. — The large, thick-ribbed, yellowish green cocoon, 
shaped, as Staudinger observes, something like a pitcher, is sus- 
pended from the twig of a tree by a stout silken thread, which 
Staudinger says measures from 10 to 30 millimeters in length. 
There is a hole at the bottom of the cocoon ; and the top, which 
is slit longitudinally, opens to pressure. It mimics a pendant 
leaf, and when the leaves have fallen from the trees in winter 
it is quite a conspicuous object, easily perceived by any person 
walking through the wpods. The larva and the pupa, as also 
stated by Jordan and Nawa, make a chirping noise, which some- 
what resembles the creaking of a bough; and the sound made 
by the pupa, so I am informed, can be distinctly heard at some 
little distance even in the woods. It makes this sound on being 
touched, but it also appears to make it of its own accord, when 
suspended by its cocoon in the woods. 

The Japanese names given to this moth are based upon the 
shape of the cocoon: Yama-bishaku, or mountain ladle; yama- 
kamasu, or mountain straw bag; yama-Ukn, or mountain basket. 

Local distribution, — Honshu, Musashi Province, Yokohama, 
November and December (Pryer) : Musashi Province, Tokyo, 
taken and bred in October and November (Wileman) : Settsu 
Province, Kobe, bred in October (Wileman). Matsumura re- 
cords the species from Hokkaido and Honshu. 

216 The Philippine Journal of Science 1921 

Time of appearance. — Ovum, April; larva, April, May, and 
June; pupa, June and July; imago, October, November, and 
December. Single-brooded. 

General distribution. — Japan; eastern Siberia (Amurland, 
Ussuri, Suifun, Vladivostok, Askold). {Jordan.) 

The type of Rhodinia fugax, a male in the British Museum 
collection, is from Yokohama. (Jonas.) 

Genus CALIGULA Moore 

Caligula Moore, Trans. Ent. Soc. London 1 (1862) 321. 
Caligula boisduvali Eversmann. 

Plate 1, fig. 10, full-grown larva, fifth ? stage; fig. 11, food plant. 

Plate 2, fig. 1, young larva, third ? stage; fig. 2, food plant; fig. 
3, thoracic segments, dorsal aspect; figs. 4 and 5, median and anal 
segments, dorsal aspect. Larva of Caligula jonasi Butler. 

Japanese name, hime-yama-nai. 

Saturniu boisduvali Eversmann, Bull. Mosc. 3 (1846) 83, pi. 1, fig. 
1; 7 (1847) pi. 4, fig. 5; Herr-Schaff., Schmett. Eur. 6 (1849) 
figs. 148-150; Oberthur, Ann. Soc. France (1886) 46 (larva); 
Staudinger, Rom. Mem. Lep. 6 (1892) 325; Staudinger and 
Rebel, Cat. Lep. Pal. 1 (1901) 127, No. 1031; Leech, Trans. Ent. 
Soc. London (1898) 265, No. 6; Matsumura, Cat. Insect. Jap. 
(1905) 47, No. 387; Thousand Insects of Japan [Nihon Senchu 
Dzukai (Jap.)] (1909) suppl. 1, 41, No. 67, pi. 6, fig. 3, 2. 

Caligula jonasi Butler, Ann. & Mag.' Nat. Hist. IV 20 (1877) 479; 
111. Typ. Lep. Het. 2 (1878) 16, pi. 25, fig. 2, d"; Leech, Proc. 
Zool. Soc. London (1888) 633, No. 252; Staudinger, Rom. Mem. 
Lep. 6 (1892) 325; KiRBY, Cat. Lep. Het. (1892) 761; Staudinger 
and Rebel, Cat. Lep. Pal. 1 (1901) 127, No. 1031a; Nawa, Insect 
World [Konchu Sekai (Jap.)] 10 (1906) 191, pi. 5, imago 2; 
Sasaki, Insects Injurious to Japanese Trees [Nihon Jiimoku 
Gaichuhen (Jap.)] ed. 3 (1910) pt. 2, 84, pi. 92, larva, pupa, imago; 
Jordan, Seitz's Macrolep. Faun. Pal. 2 (1911) 218, pi. 32, 6, c^. 

Caligula fallax Jordan, Seitz's Macrolep. Faun. Pal. 2 (1911) 217; 
pi. 32, d, 2 (as boisduvali) (subspecies jonasi = Stgr. pt.; boisduvali 
auct. pt). 

Caligula boisduvali Jordan, Seitz's Macrolep. Faun. Pal. 2 (1911 217. 

Two larvae are figured. The young larva (Plate 2, fig. 1) was 
probably in the third ? stage, but no note was made of this at 
the time. This larva was taken at Hakodate, Hokkaido (Yezo), 
in June, 1902 (figured, June 24), on wild pear, yama-nashi 
(Piru^ sp.). 

The full-grown larva (Plate 1, fig. 10) was probably in the 
fifth ? stage, but no note was made of this at the time. This 
larva was taken in June, 1901, at Yoshino, Yamato Province, 
Honshu (figured, June 17), on dwarf oak, kunugi {Quercus ser- 
rata Thunb.). The pupa of this larva died, but I bred other 

19.2 Wileman: Japanese Lepidoptera, VI 217 

imagoes of Caligula jonasi from similar larvae on September 21, 
29, and 30, and October 1, at Hakodate. Pryer states that *'the 
larva feeds on cherry and that it resembles a small Caligula 

Staudinger comments on the larva of Saturnia boisduvali 
Eversm. ( ? var. jonasi) , as follows : 

The yellow green (preserved) larvae are covered with very short, bristly, 
yellow hairs, with a few, long, dark hairs, especially on the anterior seg- 
ments. When young they have a broad, black, dorsal stripe with orange 
warts. They appear to differ a great deal, for Oberthiir describes them 
in Ann. Soc. Fr. (1886), p. XLVI, as "black, with two red dorsal spots." 
They make a light lattice-like cocoon which resembles those made by the 
larvae of Saturnia caecigena, with which they have some resemblance, and 
the moth appears, as in the case of caecigena, in autumn (September). 

The six, paired, dorsal spots of my young larva of Caligula 
jonasi are red as described by Oberthiir, not orange as described 
by Staudinger,^^ who, possibly, is speaking of the Amurland 
geographical form, subspecies falUix Jordan (not named by 
Jordan at the time the larva was described by Staudinger) . 

Nawa ^^ states that — 

Jonasi hibernates in the ovum stage and the young larva emerges in 
May. The full-grown larva spins up and pupates, after the fourth ? or 
fifth ? molt is completed, about the middle of June and the imago emerges 
about the middle of October, when the female oviposits. 

The larva feeds on nashi, pear IPirus sinensis Lindl.] ; ume, plum 
[Prunus mume S. and Z.] ; keyaki [Zelkowa acuminata Lindl.] ; midzuki 
[Comus macrophylla Wall]. At first it is black with tubercles on each 
segment which emit black and gray hairs; legs black; prolegs grayish 
yellow; when it grows in size the head becomes deep green and the body 
light green, dorsally tinged with white; the subspiracular line becomes 
light green; and the ventrum grayish green; spiracles chestnut color; 
the hairs of the subspiracular series of tubercles very long when full- 

Oberthur gives as food plants in Manchuria [? Amurland] 
Betula; Prunus; Pirus baccata Linn, [found and known in 
Japan as var. manchurica Maxim., under the Japanese names 
of ko-ringo, beniringo, aka-ringo], Sasaki ^^ gj^gg descriptions 
and figures of the young and adult larva, cocoon, and male imago 
of Caligula jonasi. He says : 

The larva emerges about the month of June and feeds upon onara, 
oak [Quercus crispula Blume] ; gamazumi, [Viburnum dilatatum Thunb.] ; 

'° Rom. Mem. Lep. 6 (1892) 326. 
"^ Insect World (Konchu Sekai) 10 (1906) 191. 

''Insects Injurious to Japanese Trees [Nihon Jumoku Gaichuhen (Jap.)] 
ed. 3 (1910), pt. 2, 34, pi. 92. 

218 ^'^^ Philippine Journal of Science 1921 

kurumi, walnut [Juglans sp.]. It is full grown between September and 
October and the imago emerges by the end of October. 

I describe the larva from' my original figures as follows : 

Young larva in third ? stage, — Plate 2, fig. 1. Length, 23 
millimeters ; light green ; a broad diamond-pattern, longitudinal, 
black dorsal stripe with paired red tubercles emitting short, 
spinous black hairs on segments 3, 4, 7, 8, 9, and 10 (counting 
head as segment 1), paired black tubercles, with similar hairs 
on segments 2, 5, 6, 11, 12, and 13; subspiracular line lighter in 
color; head pale green, edged with black; spiracles dark; legs 
dark; prolegs and claspers green. 

Larva in fifth ? stage.— Pldite 1, fig. 10. Length, 65 milli- 
meters. Body yellowish green, head of a darker shade ; dorsum 
covered with short, bristly, yellow hairs with a few long, darker 
hairs interspersed amongst them; segmental sutures whitish; 
spiracles white, encircled with a ring of chestnut brown; sub- 
spiracular line yellowish; legs ochraceous; prolegs dark green; 
claspers yellowish green, with a long black dash at the base. 

My original figure of the larva in the third ? stage agrees 
with that given by Sasaki. Jordan comments on Caligula hois- 
duvali as follows: 

Larva green, above and below with a black-brown, longitudinal stripe, 
the warts of both dorsal rows reddish yellow, full-grown without stripes 
(always ?). On various deciduous trees. Cocoon reticulate. The moth 
in the autumn. From Lake Baikal to Japan in three geographical forms. 

Boisduvali Ersch. [ = Eversm.], from Kiachta, Urga and the Kentei Moun- 
tains, southward of Lake Baikal * * *. 

Fallax [Jordan], subsp. nov. (Jonasi Stgr. pt.; boisduvali auct. pt.) (32d 
as boisduvali). * * * Distributed from Vladivostok, Askold, Ussuri to 
the Amur. 

Jonasi Butler (32b, d). * * * Japan, on the Main Island, [Honshu] 
in September. Two nearly full-grown larvae before me have no longi- 
tudinal stripe above or below. 

Cocoon. — ^The pupa is enclosed in a cocoon similar to that of Dictyoploca 
japonica but it is smaller and spun with finer mesh. It also resembles, 
as Pryer observes of the cocoon of japoncia, "the wire net in a cartridge." 

Local distribution. — Honshu, Musashi Province, Yokohama 
(Pryer) : Musashi Province, Tokyo, October and November 
(Wileman) : Shinano Province, Oiwake, September and October 
(Leech) : Shimotsuke Province, Nikko, September and October 
(Leech), Hokkaido (Yezo), Oshima Province, Hakodate, Sep- 
tember and October (Wileman). Matsumura records Saturnia 
(Caligula) boisduvali from Honshu and Hokkaido. 

19,2 Wileman: Japanese Lepidoptera, VI 219 

Time of appearance. — Larva, May and June; pupa, June to 
September ; imago, September to November ; ovum, November to 

General distribution. — From Lake Baikal to Japan. (Jor- 


Dictyoploca Jordan, Seitz's Macrolep. Faun. Pal. 2 (1911) 218. 

Dictyoploca japonica Butler. 

Plate 2, fig. 6, full-grown larva. 

Japanese names: Kuso-san, tegusu-ga, kuri-kemushi, tsuzuri-no-nishikiy 
shiragata-ro, kurimitshi-ga. 

Caligula japonica MooRE, Trans. Ent. Soc. London (1862) 322 (pupa 
case) (regina Stgr.) ; Butler, Ann. & Mag. Nat. Hist. IV 20 (1877) 
479; 111. Typ. Lep. Het. 2 (1878) 16, pi. 26, fig. 2, c?; Pryer, Trans. 
Asiat. Soc. Japan 12 (1883) 52, No. 191; Oberthur, Ann. Soc. Ent. 
France (1886) 48, larva; Leech, Proc. Zool. Soc. London (1888) 
633, No. 253; Staudinger, Rom. Mem. Lep. 6 (1892) 328; Leech, 
Trans. Ent. Soc. London (1898) 264, No. 5; Matsumura, Injurious 
Japanese Insects [Nihon Gaichuhen (Jap.)] (1899) 75, pi. 32, fig. 
1, imago d*; Staudinger and Rebel, Cat. Lep. Pal. 1 (1901) 126, 
No. 1026; PACKARD, Proc. Am. Acad. Arts and Sci. 39 (1904) 
564 (larva); Matsumura, Cat. Insect. Jap. 1 (1905) 47, No. 385; 
NAWA, Insect World [Konchu Sekai (Jap.)] 10 (1906) 63, pi. 3, 
larva, pupa, imago d*; Matsumura, Thousand Insects of Japan 
[Nihon Senchu Dzukei (Jap.)] (1909), suppl. 1, 50, No. 82, pi. 9, 
fig. 1, ?; Sasaki, Insects Injurious to Japanese Trees [Nihon 
Jumoko Gaichuhen (Jap.)] ed. 3 (1910) pt. 2, 89, pi. 119, larva, 
pupa, imago c^; Sasaki, Insects Injurious to Fruit Trees [Kv^raju 
Gaichuhen (Jap.)] ed. 5 (1911) 216, pi. 70, larva, imago. 

Caligula castanea Sv^^iNHOE, Cat. Lep. Het. Oxford (1892) 249; 
Jordan, Seitz^s Macrolep. Faun. Pal. 2 (1911) 218, pi. 32 c, c^. 

Dictyoploca japonica Jordan, Seitz's Macrolep. Faun. Pal. 2 (1911) 
218, pi. 32, c, c^, ?. 

Jordan erects a new genus, Dictyoploca, for this species, 
which was formerly placed in Caligula Moore. 


Forewing v^^ith only tw^o subcostal branches. The distal segments of the 
antennae below^ more produced at the apex than in Caligula, and v^ith more 
distinct sensory cones, the pectinations of d* shorter, the apical ones on 
the middle segments of the 2 short but distinct. Larva at first black, 
with rows of warts clothed with black bristles, in the following stages: 
greenish yellow below, black above, clothed with long white hairs, then 
more or less completely white with blue spots at the stigmata, small black 
dots, and short transverse streaks laterally. Cocoon with much larger 
meshes than in Caligula. Pupa very rugate, apex of abdomen almost 

220 ^^^ Philippine Journal of Science 1921 

truncate in a straight line, forming a sharp edge, on each side with a 
bundle of short bristles in a groove. The cocoons are used for the manu- 
facture of silk, but both the quantity and quality are negligible. Himalayas 
to Japan. 

Jordan '^ gives the following details concerning the imago 
and larva of Dictyoploca japonica: 

D. japonica Butl. (regina Stgr.) (32c). Ground-colour varying, yellow- 
ish grey, brownish yellow, or almost olive; the markings on the contrary 
fairly constant. Forewing above in the basal fourth with a reddish or 
brownish transverse line which bounds a large basal spot, and with a second 
rather diffuse line beyond the middle which touches the ocellus at the outer 
side, or stands slightly distant from it; the area bounded by the two lines 
slightly lighter than rest of the wing; ocellus oblique. Hindwing usually 
redder than the forewing, with much larger and more sharply defined ocellus. 
Below more unicolorous than above; the ocellus on the forewing with 
black pupil, on the hindwing blind. Pale specimens are f. castanea Swinh. 
Larva on Juglans, Castanea, Camphora, in captivity takes Oak, Hawthorn, 
Willow, etc. The three first stages of the larva almost alike; in the fourth 
and fifth stages the black colour confined to the sides, the warts of the 
thorax with a few black bristles between the greenish white hairs. Meshes 
of cocoon large. Japan (Main Island and Kyushu), Amurland, North 
China; the moth in the autumn (September and October), common. The 
silk glands of the caterpillar are sometimes employed for the manufacture 
of fishing lines. 

The type of D, japonica, male, from Yokohama (Jonas) is in 
the British Museum collection. The larva figured (Plate 2, 
fig. 6) was taken in June, 1901, at Kobe, Settsu Province, Hon- 
jshu, on kunugi, dwarf oak (Quercus serrata Thunb.). It pu- 
pated on June 27, and a female imago of the pale buff form 
emerged on a date unrecorded. 

Packard ^* has described the larva at length, giving all the 
stages. Nawa ^^ and Matsumura ^^ record the life history and 
give figures of the ova, larva, pupa, cocoon, and male imago. 
Oberthur ^^ also has described the larva. 

Food plants, — The food plants are as follows, and probably 
several more should be included: Ku-su-no-ki, camphor tree 
(Cinnamomum camphora Nees) ; kurumi, walnut (Juglans 
sp.) ; ringo, apple, (Pyrus malus L.) ; uru^hi, lacquer tree (Rhus 
vemicifera DC.) ; hakuyo, poplar (Populas alba L.) ; kusagi 
(Clerodendron tricotomum Thunb.). The larva, however, is 

^'Seitz's Macrolep. Faun. Pal. 2 (1911) 218. 

''Proc. Am. Acad. Arts Sci. 39 (1904) 564. 

"Insect World (Konchu Sekai) 10 (1906) 63, pi. 3. 

'* Japanese Injurious Insects (Nihon Gaichuhen) (1899) 75, pi. 32. 

'' Soc. France (1886) 48. 

19,2 Wileman: Japanese Lepidoptera, VI 221 

found chiefly on kuri, Spanish chestnut (Castanea vulgaris Lam. 
var. japonica DC), whence its popular name of kuri-kemushi or 
kiiri-rmishi is derived, equivalent to "chestnut caterpillar'' or 
"chestnut grub'' in English. Oberthtir gives Jtiglans mantsch- 
ourica as the food plant in Amurland. 

Cocoon. — Pryer ^^ remarks : 

Commonly called the wire-cartridge moth, from the resemblance of the 
cocoon to the wire-net in a cartridge; last year, [1884], it was found feed- 
ing on poplars newly introduced into this country [Japan]. The natives 
make a strong coarse silk from the cocoon, and a fine gut from the intestines 
of the larva. Imago appears in October; larva hairy. 

Nawa ^^ gives the following description of the larva which, in 
his figure, measures about 60 millimeters, much smaller than 
mine, which measures 103 millimeters. 

The larva at first is black and the head is covered with ashy white 
hairs; six small tubercles on each segment which emit long ashy white 
and black hairs; these become pale green when the larva grows bigger; 
head greenish yellow covered with pale yellow hairs; body pale green; 
the tubercles of each segment of the body emit long greenish white hairs, 
for this reason it is named shiragnta-ro ; spiracular stripe black;, spiracles 
indigo-blue; subspiracular line yellow spotted with light red; ventrum 
dark yellowish green speckled with black spots; legs pale yellow; prologs 
dark yellowish green and their extremities pale yellowish green. The 
cocoon is woven like a net and is commonly called sukashi-dawaray [or 
transparent rice-bag, because the pupa can be seen through the meshes.] 

Sasaki ^^ gives descriptions and figures of the larva, pupa, 
and male imago of Dictyoploca japonica. He says : 

* * * the larva emerges between the end of April and commence- 
ment of May and feeds upon hyakujikko, or sarusuberi [Lagerstroemia 
indica Linn.] ; kuri, [Spanish chestnut {Castanea vulgaris Lam. var. japo- 
nica DC.]; kusu-no-kij [Camphor tree (Cinnamomum camphora Nees)]; 
[keyaki (Zelkowa acuminata PI.)]; kurumi\ [walnut (Juglans sp.)]. It is 
full-grown between the middle of June and commencement of July and the 
imago emerges between the end of August and the commencement of 
October, when it oviposits. 

Matsumura -^ records the life history of the species and gives 
figures of the male imago, ova, larva, and cocoon. He says that 
''in Hokkaido it is single brooded and hibernates in the ovum 
stage." The female deposits some three hundred forty or more 

''Trans. Asiat. Soc. Japan 12 (1883) 52. 

''Insect World 10 (1906) 63. 

'"Insects Injurious to Japanese Trees [Nihon Jumoku Gaichuhen (Jap.)] 
ed. 3 (1910) pt. 2, 89, pi. 199, larva, pupa, imago; Insects Injurious to Fruit 
Trees [Kwaju Gaichuhen (Jap.)] ed. 5 (1911) 216, pi. 70, larva, imago. 

''Japanese Injurious Insects (Nihon Gaichuhen) (1899) 75. 

222 The Philippine Journal of Science 1921 

ova in two or three different places on the bark of the tree on 
which it feeds. The ova hatch in May or June of the following 
year. At first the larva is black; after the second molt white 
hairs begin to appear, and it is full grown in about forty-five 
days, when it makes a netlike cocoon, which is commonly known 
as sukashi-dawara. The imago emerges at the end of September, 
when the ova are deposited by the female in the manner pre- 
viously indicated. The larvae sometimes appear in such num- 
bers that they do much injury to the foliage of chestnut and 
walnut trees, but this is profitably counterbalanced to some 
extent by the fact that a kind of catgut named tegusu, used by 
fishermen, is made out of threads contained in the body of the 
full-grown larva and finds a ready sale. 

Local distribution, — Honshu, Musashi Province, Yokohama 
(Pryer) ; Tokyo, taken and bred September and October 
(Wileman) : Shimotsuke Province, Nikko, October (Leech) : 
Settsu Province, Kobe, September (Wileman) , Kyushu, Kyushu 
(Leech) : Hyuga Province, Kuraoka, September (Wileman). 
Shikoku, my Japanese collector, informed me that he had taken 
the species in Sanuki Province. Matsumura records it from 
all the Japanese Islands, Hokkaido, Honshu, Kyushu, Shikoku, 
and from Formosa (Taiwan). 

Time of appearance, — Ovum, September and October; larva, 
April to June; pupa, June to September; imago, September and 
October. The imago emerges in September and October and 
then the female deposits her eggs on the bark of the trees and on 
twigs, which pass the winter without hatching. The young 
larvae emerge in May of the following year. Single-brooded. 

As has been indicated by authors previously quoted, a gut, 
named in Japanese tegusu-ito, is manufactured from the silk 
glands or the intestines of the full-grown larva. Those in- 
terested in this subject and in the manufacture of the coarse 
silk, also made out of the strong netlike cocoon, can find further 
details in several Japanese scientific journals." 

The following account of the manufacture of tegusu-ito is 
given by Sasaki.-'^ He quotes an extract from a pamphlet, or 

"Toba, Insect World (Konchu Sekai) 1 (1897) pt. 3, 92-94; Sasaki, 
Nihon Jumoku Gaichuhen (Insects Injurious to Japanese Trees) ed. 3 
(1910) 92; Sasaki, Dai Nihon Nokwaiho (Bull. Jap. Agr. Soc.) (1905) 
No. 294, 15-18; Honda, Dai Nihon, Sanshi Kwaiho (Bull. Jap. Sericult. 
Soc.) (1905) No. 160, 1-7; Honda, Dai Nihon Nokwaiho (Bull. Jap. Agr. 
Soc.) (1905) No. 292, 3-11. 

^' Nihon Jumoku Gaichiihen. 

19,2 Wileman: Japanese Lepidoptera, VI 223 

article, named *'Kusumushi Yo ho/' **the method of rearing 
camphor grubs," which is as follows: 

Many persons have tried to manufacture "tegusu-ito" from the full- 
grown larvae of the "kuri-mushi" (chestnut grub). The process of manu- 
facture is as follows; 

Full-grown larvae which are just on the point of spinning their codoons 
are selected. They are slit open and the silken threads extracted. These 
threads are then soaked in a solution made of vinegar, (of 35 per cent 
strength), pure water and a little salt. After that they are removed 
from the solution, stretched out and dried, and are now called "tegusu-ito.'^ 
These threads are now soaked again in white rice water [which is the 
water resulting after washing rice, preparatory to cOoking], for about one 
hour, after which they are taken out and dried. When thoroughly dry 
they are placed between walnuts wrapped up in layers of cotton cloth and 
gently rubbed. This imparts to them gloss and elasticity and a fine-looking 
**tegusu-ito" is the result. 

It is to be presumed that fresh green walnuts are used, as 
without the outer green skin they would probably be too hard 
and uneven for this purpose. 

Genus AGLIA Ochsenheimer 

Aglia Ochsenheimer, ScTimett. Eur. 3 (1810) 11. 

Aglia tau Linnaeus. 

Plate 2, fig. 7, larva; fig. 8, head; fig. 9, extended transverse section of 
larva; fig. 10, food plant. Larva of Aglia japonica Leech. 

Japanese name, yezo-yotsume. 
Bombyx tau Linnaeus, Syst. Nat. 1 (1758) 497; Hubner, Bomb. 
(1800) pi. 13, figs. 51, 52; Staudinger and Rebel, Cat. Lep. Pal. 1 
(1901) 127, No. 1039; Jordan, Seitz's Macrolep. Faun. Pal. 2 (1911) 
224, pi. 35, a, d",?; 35, 6,?. 

Aglia tau var. japonica Leech, Proc. Zool. Soc. London (1898) 632, 
No. 250; Trans. Ent. Soc. London (1898) 269, No. 16; Staudinger 
and Rebel, Cat. Lep. Pal. 1 (1901) 127, No. 1039 c; Matsumura, 
Cat. Insect. Jap. 1 (1905) 47, No. 388; Thousand Insects of Japan 
[Nihon Senchu Dzukai (Jap.)] (1909) suppl. 1, 49, pi. 8, fig. 
5, c?; Jordan, Seitz's Macrolep. Faun. Pal. 2 (1911) 224, pi. 35, a, d". 

The larva figured (Plate 2, fig. 7) was taken in August, 1902 
(figured, August 18), at Hakodate, Oshima Province, Hokkaido, 
on maple, Japanese name momiji (Acer sp.). A female imago 
of var. japonica Leech emerged from the resulting pupa on 
July 12, 1903. 

Jordan ^^ describes the larva, pupa, and cocoon of the genus 
Aglia as follows: 

Young larvae with two dorsal rows of warts, which are long and thorn- 
like on the pro- and mesothorax, on segment 11 only one such thorn, anal 

''Seitz's Macrolep. Faun. Pal. 2 (1911) 224. 

224 ^^^ Philippine Journal of Science 1921 

segment with a median thorn at the tip; laterally and dorsally numerous 
granules bearing small hairs; in the later stages the warts completely 
obsolescent, but the segments swollen transversely at the corresponding 
places. Cremaster of pupa sharp with curved bristles. 

Cocoon loose, between leaves and moss or below the surface. 

Jordan ^'' describes the larva of Aglia tau as follows : 

Larva green vdth light lateral stripes directed obliquely up- and back- 
ward, below the stigmata a light longitudinal line with a reddish edge, the 
line widened on segment 4 to form a black-centered spot. On deciduous 
trees especially Beech, Oak, Birch, etc. Pupa hibernates. Moth from 
March to June according to the locality, in the North later than in ^le 
South; the c?c? fly by day and are very restless, the ?? remain on tree- 
trunks and on the ground. In the Central and Southern districts of 
Northern Europe, eastward to Japan; not in England and the Mediter- 
ranean countries. 

Kirby ^^ remarks : 

The larva [of Aglia tau] is green, with five red spines when young, 
which it loses when full grown. It has yellowish-white oblique stripes on 
the side, running upwards and forwards, and a yellowish line on the sides, 
which is broadest on the 4th segment. It feeds on beech, lime and oak in 
June and July. 

The original figure of my larva of var. japonica Leech agrees 
well with the description given by Jordan of Aglia tau Linn., 
but he does not give the number of the oblique lateral stripes 
which, in my larva, are seven in number and of a whitish color. 

Poulton 27 remarks : 

The larva of the European Tau Emperor (Aglia tau) has an eye-like 
mark which it can expose when attacked, but which is otherwise concealed. 
The appearance of the larva in its terrifying attitude is shown in fig. 58. 

This larva is an example of the form of protective mimicry 
alluded to by Poulton under pseudaposematic colors.^^ In the 
figure of my larva (Plate 2, fig. 7) the eyelike spot on segment 
5 does not seem to be fully expanded. 

Local distribution of Aglia tau var. japonica. — Honshu, Shi- 
motsuke Province, Nikko, May (Wileman). Hokkaido, Oshima 
Province, Hakodate and Jansai Numa, June and July (Wileman) . 
Matsumura records var. japonica from Honshu and Hokkaido. 

Time of appearance. — Larva, August; imago. May to July. In 
Hokkaido the larva pupates in September, and the imago emerges 
in July of the following year. 

''hoc. cit. 

^European Butterflies and Moths (1889) 125. 

''The Colours of Animals (1890) 264, fig. 58. 

-^'Vide Philip. Journ. Sci. § D 9 (1914) table 1, facing page 248. 

19, 2 Wileman: Japanese Lepidoptera, VI 225 

General distribution. — Aglia tau, in the central and southern 
districts of northern Europe, eastward to Japan; not in Eng- 
land and the Mediterranean countries. Aglia tau var. japonica, 
Japan only. {Jordan.) 

Genus BRAH]«LaSA Walker 
Brahmma Walker, Cat. Lep. Het. 6 (1855) 1315. 
Brahmaea japonica Butler. 

Plate 2, fig. 11, larva, fourth ? stage; fig. 12, food plant; fig. 13, head; 
fig. 14, head, enlarged. 

Japanese names: Ibota-ga and shokko-nishiki. 

Brahmsea japonica Butler, Ent. Month. Mag. 10 (1873) 56 {mnis- 
zechii Feld.); 111. Typ. Lep. Het. 2 (1878) 17, pi. 26, fig. 3, 6; 
Pryer, Trans. Asiat. Soc. Japan 12 (1883) 53, No. 194: Leech, 
Proc. Zool. Soc. London (1898) 635, No. 257; Trans. Ent. Soc. 
London (1898) 270, No. 20; Matsumura, Cat. Insect. Jap. (1905) 
48, No. 389; Nawa, Insect World [Konchu Sekai (Jap.)] 10 (1906) 
415, pi. 11, larva, pupa, imago ?; Matsumura, Thousand Insects of 
Japan [Nihon Senchu Dzukai (Jap.)] (1909) suppl. 1, 42, No. 
68, pi. 6, fig. 4, 2; Sasaki, Insects Injurious to Japanese Trees 
[Nihon Jumoku Gaichuhen (Jap.)] ed. 3 (1910) pt. 3, 132, pi. 225, 
larva, pupa, imago; Jordan, Seitz's Macrolep. Faun. Pal. 2 (1911) 
228, pi. 35, fig. c, (?. 

Brahmsea mniszechii Felder and Rogenhofer, Reise Novara, Lep. 4 
(1874) pi. 93, figs. 4, 5. 

Brahmaea nigrans Butler, Ent. Month. Mag. 17 (1880) 110; Water- 
house, Aid 1 (1881) pi. 29. 

Jordan ^^ comments on the family Brahmseidae as follows : 
This group comprises about one and a half to two dozen of highly 
peculiar but very similar species. All are large and rather clmnsy moths, 
with markings so characteristic that they at once catch the eye even in 
large and mixed collections. The wing is divided into an outer half trav- 
ersed by ten parallel wavy lines, which on the forewing directly touches 
an often uniformly dark basal area, but on the forewing borders on a 
band which is sometimes modified at the inner margin to form an ocellus- 
like disc. The basal area of the forewing again contains a number of 
those peculiar parallel lines, which renders the *scheme of markings so con- 
fusing, and the biological significance of which we do not yet understand. 
And as if even Nature could not carry out so complicated a pattern in 
all its details, we very often find among the Brahmaeidae unsymmetrical 
specimens in which one side bears sometimes one stripe more than the 
other, sometimes has the dots differently placed. Among a considerable 

"•Seitz^s Macrolep. Faun. Pal. 2 (1911) 227. 

180732 7 

226 The Philippine Journal of Science 1921 

number of otherwise very well developed B. japonica before me there is 
not one specimen the two sides of which agree in every detail. 

The position of Brahmaea in classification has only changed in that 
it was sometimes placed among the Satumiids, sometimes among the Bom- 
bycids. The larvae as far as they are known, when full-grown resemble 
huge larvae of silk-moths but differ greatly from the latter in the early 
stages, which shows that there is no close relationship. It will be absolutely 
necessary to keep the genus Brahmaea separate as a distinct family, and 
it is very noteworthy that a similar phenomenon exists in America, a very 
homogeneous family in many respects resembling Brahmaea also stand- 
ing alone nearly without transitions and playing the same part in the 
fauna of the New World as Brahmaea in the Old World. These are 
the Ceratocampids, the largest species of which is produced from that 
grotesque and strange caterpillar, with its curved horns on the thorax 
which we figure on the cover of each part of this work in a defensive 
attitude and which perhaps those who are not familiar with the Amer- 
ican fauna may have considered a product of the imagination. 

The Brahmaeids are confined to the Old World and are so distributed 
that three species occur in the Palearctic Region, but not in Europe, 
just as many forms are Indian, and the same number belong to the 
Ethiopian fauna. They do not go far north and inhabit mountainous 
countries, have only one brood in the temperate zone, and as larvae are 
fairly polyphagous. The larvae grow slowly and pupate in the ground 
without a cocoon; the moths fly at night, and rest by day on tree-trunks 
and branches, where, with their wings in steep roof-shape and folded close 
together, they resemble fruits or pieces of bark. The moths are rather 
rarely seen, but the larvae are common wherever they occur, and lately 
large quantities of material for breeding have been imported. 

The characters of the family are those of the single genus Brahmaea. 

* * * the larvae, when young, with long horns decreasing in length 
when the larva grows older, otherwise naked, soft, long and not strong; 
on deciduous trees. 

The larva figured (Plate 2, fig. 11) was taken in August, 1902 
(figured, August 2), at Hakodate, Oshima Province, Hokkaido 
(Yezo), on ibotanoki (Ligtcstrum ibota Siebold), but died with- 
out reaching the pupal stage. It is so well known amongst 
Japanese collectors that, although I have never succeeded in 
breeding the imago, I have no hesitation in referring the larva 
figured to Brahmsea japonica, Pryer ^^ moreover comments on 
the larva as follows: 

Feeds on the privet; larva is smooth, bright green, marked with black, 
and has four thin black filaments over 11 inches long on the foresegments 
and three on anterior [posterior]. Imago appears in March and April. 

Pryer, however, speaks of the larva previous to the fifth molt 
when it loses all these filaments. My figure represents the larva 
in the fourth ? stage. 

•"Trans. Asiat. Soc. Japan 12 (1885) 53. 

19, 2 Wileman: Japanese Lepidoptera, VI 227 

Nawa ^^ records the life-history of the species and gives figures 
of the larva in its fourth ? and fifth ? stages with and without 
these curious filaments which are seven in number, also of the 
pupa and of the female imago. The following description is 
taken from my original figure : 

Larva. — Fourth ? stage. Length, about 60 millimeters. 
Head black with a pattern of yellowish markings. Color pale 
whitish green, tinged with yellowish green patches subdorsally 
and laterally ; a midlateral and spiracular series of black spots, 
streaks, and dashes; a suprapedal series of black elongated 
dashes situated on yellow patches; legs, prolegs, and claspers 
black, prolegs and claspers being dotted or streaked with white ; 
ventrum blackish; segment 3 (counting head as segment 1) 
bears two black, wirelike filaments, much contorted, about 26 
millimeters long; segment 4, two similar filaments of about 
equal length ; segment 12 bears one similar, shorter, mediodorsal 
filament; the anal segment bears two filaments which are the 
shortest of the seven ; all the filaments are more or less contorted 
and vary in length with the individual larva. The full-grown 
larva after the fifth molt measures from 90 to 110 millimeters and 
changes to yellowish brown. Nawa states that when it reaches 
maturity it loses these seven filaments and that it burrows in the 
earth where it pupates. He does not say, however, whether it 
forms a cocoon in the earth. My Japanese collector informed 
me that it makes a hard, subterranean cocoon, but I have never 
investigated the matter and am therefore unable to say whether 
it forms a cocoon or not. Nawa further says : 

* * * there is only one brood in the year. The imago emerges in 
April /[at Gifu, Omi Province], the ova hatch in May, the larvae appear 
in May and June and the pupa passes the winter under ground until the 
following April. The larva feeds on nedzumw-mochi [{Ligustrum japonica 
Thunb.)] and ibotanoki {(Ligustrum ibota Siebold)]. 

Ibotanoki is also called yachi-tamo in Hakodate, Hokkaido. 
Sasaki ^^ gives descriptions and figures of the larva, pupa, and 
male imago of Brahmsea japonica. He says: 

♦ * * the larva emerges about the middle of May and feeds upon 
nedzumi-mochi and ibotanoki. It is full grown by the middle of June 
and then pupates in the earth. It hibernates in the pupal stage and 
the imago emerges in April of the following year. 

"Insect World (Konchu Sekai) 10 (1906) 415, pi. 11. 
"Insects Injurious to Japanese Trees [Nihon Jumoku Gaichuhen (Jap.)] 
ed. 3 (1910), pt. 3, 132, pi. 225. 

228 The Philippine Journal of Science 1921 

I may mention that the larva of this species finds a place in 
the materia medica of the lower class Japanese as a specific for 
certain diseases, and Nawa corroborates me in this statement. 
He says : 

Popular superstition attributes to this larva some efficacy in cases of 
consumption but I have never heard of any cures effected in such cases. 

I have often in my wanderings through the Bukenji Woods, 
near Yokohama, met men carrying shallow baskets filled with 
these larvae and, on inquiry, have learned that they were to be 
sold to native apothecaries. They are sold by collectors at 
about 3 sen (1.5 cents gold) per larva, and after being dried in 
the sun are grilled over a fire. The retail price at which they 
are sold by the apothecary ranges from 4 to 5 sen per larva 
(2 to 2.5 cents gold). It has been stated to me that these dried 
larvse are supposed to be efficacious in cases of piles, convulsions, 
and worms in children. As a matter of fact the lower class 
Japanese employ the larvse of several other species of Heterocera 
for similar purposes, such as those of Sciapteron regale Butler 
wich feeds inside the stems of the wild grape, yama-budo (Vitis 
coignetix Pull.), and also of Phassus signifer Walker and Phas- 
sus excrescens Butler, which are found in the stmes of the kusagi 
(Clerodendron tricotomum Thunb.) and other trees. This how- 
ever is not surprising when one considers that the Romans are 
reported to have regarded the larva of Cossics ligniperda L., 
also an internal feeder, as a great table delicacy. 

The following description of the larva of the allied species 
Brahm^a certhia Fabr. {=undulata Brem. and Grey; petiveri 
Butler) , from Amurland, northern and central China, and Korea, 
is given by Seitz and is useful for comparison with that of B. 
japonica, which it seems to resemble in some respects : 

Larva grey to blackish, when young with two horns rolled up at the 
ends on segment 2 and two similar ones on 3; the remaining segments 
only have small knobs, the anal segment bears an also very strongly 
curved horn resembling that of the Sphingidae. When full grown the 
larva is smooth, segments 2 and 3 swollen, on the anal segment a stmnpy 
hump. Markings and colour very variable, frequently confined to a few 
small black streaks or bright-coloured spiracular dots. According to Staud- 
inger the larvae of the East- Asiatic form are much more brightly marked 
and coloured than the almost uniformly dark larvas of ledereri of Asia 
Minor [a form of certhia Fab.] . According to Korb the larvae often live in 
companies of from 20 to 30 specimens and prefer rather sunny sterile 
localities; until June or August on Privet, Syringa, Ash, Philljnrea and 
other trees; when disturbed they emit a cracking or crackling noise. They 
pupate without a cocoon in or on the ground beneath stones ; pupa blackish, 

15. 2 Wileman: Japanese Lepidoptera, VI 229 

very strongly glossy, clumsy, stumpy at both ends, deeply incised between 
the segments, rounded at the abdominal end, the cremaster a short point. 
The moth appears in June in Eastern Asia. 

Imago. — Leech ^^ remarks : 

This is a variable species both as regards ground-colour and markings. 
Thus the former may be white or grey, and sometimes tinged with green; 
then the number of ocelli in central band and "rounded internal spot" 
is not the same in any two individuals comprised in my series of 15 speci- 
mens. The spot referred to sometimes has three ocelli across its centre 
on one wing, but its companion on the other wing has four. In all cases 
the number of ocelli, both in the spot and central band, is greater on one 
side than the other. Again the shape of the central band is subject to 
modification, and stages in the formation of the rounded internal spot 
from the lower portion of this band are exhibited in the specimens in 
my Japanese series; thus between an example in which the band is entire 
from costa to inner margin, and but slightly contracted below the middle, 
and a specimen with the rounded spot completely formed and quite in- 
dependent, there are all the intermediate stages. 

The Type of Brahmsea japonica, a male ?, is from Yokohama, 
Japan. The type of B. nigrans is from Japan. 
Seitz 2* remarks as follows : 

B. japonica Butl. (mniszechii Feld.) (35c). Smaller than the preceding 
[B. christophi Stgr.], more grey and often tinged with greenish. The 
median band of the forewing is strongly widened in the costal area; 
lighter and ornamented with small rings. These are nearly always asym- 
metrical; Leech did not find one symmetrical individual among 15 in 
Fryer's collection and the same applies to four specimens collected by me. 
The original of our figure also has on the right forewing (not reproduced) 
much fewer rings than on the wing figured. The median band also 
varies rather strongly in its inner marginal portion. Now it is more now 
less constricted on the lower median vein; but nearly always it forms 
an ocellus-like disc-patch. The species also varies considerably in the 
whole scheme of colouring and nigrans Butl., unknown to me in nature, 
is probably only a dark specimen; japonica is also closely allied to the 
Indian conchifera Butl. and is only a Northern form of wallichii Gray 
(spectabilis Hope), which we figure in Vol. 10, its ground-colour shading 
into greyish green and that of rufescens Butl. into reddish. These latter 
are Indian forms, which perhaps vary slightly according to locality, but 
belong to one species. But if Leech places wallichii with certhia, he is 
in error. — Japonica is not rare near Yokohama; occurs also in other local- 
ities on Honto, [(Honshu)] and is also found in Hokkaido. 

Local distribution. — Honshu, Musashi Province, Yokohama, 
March and April (Pryer) ; Tokyo, March and April (Wileman) : 

"Proc. Zool. Soc. London (1888) 635, No. 257. 
''Macrolep. Faun. Pal. 2 (1911) 228. 

230 The Philippine Journal of Science 

Sagami Province, Hiratsuka, April (Wileman). Matsumura 
records it from Hokkaido and Honshu. 

Time of appearance, — Larva, May, June, July ? and August; 
imago, March, April, and May. In Yokohama and Tokyo, Hon- 
shu, the larva is found in May. It pupates in June and lies 
until March of the following year, the earliest emergences taking 
place about the middle of March. In fact this species is one of 
the earliest of the Japanese moths to appear in the spring. It 
is often to be taken at rest on the trunks of pines or oaks. The 
larva taken at Hakodate in August, from which my original 
figure was drawn, was exceptionally late in the season owing 
to the more northern latitude of Hokkaido. 

General distribution. — ^Japan. Is a northern form of wallichii 
Gray (spectabUis Hope). (Seitz.) 


[Drawings by Hisashi Kaido.] 

Plate 1 

Figs. 1 to 5. Samia pryeri Butler. 

i, larva, third ? stage; 2, larva, fourth ? stage; 3 and -4> 
larva, fifth ? stage, dorsal and lateral aspects; 5, larva, 
sixth ? stage; 5a, food plant. 
6 to 9. Rhodinia fug ax Butler. 

6f larva fifth ? stage; 7, larva, sixth ? stage; 8, paired 
dorsal tubercles on segment 4; 9, tubercles on segments 
11 and 12. 
10 and 11. Caligula jonasi Butler. 

10 y full-grown larva, fifth ? stage; 11, food plant. 

Plate 2 

Figs. 1 to 5. Caligula jonasi Butler. 

i, young larva, third ? stage; 2, food plant; 5, thoracic 
segments, dorsal aspect; ^ and 5, median and anal seg- 
ments, dorsal aspect. 
Fig. 6. Dictyoploca japonica Butler, full-grown larva. 

Figs. 7 to 10. Aglia japonica Leech. 

7, larva; 8, head of larva; 9, extended transverse section 
of larva; 10, food plant. 
11 to 14. Brahmsea japonica Butler. 

11 y larva, fourth ? stage; 12, food plant; IS, head; lU, 
head, enlarged. 


WiLEMAN : Japanese Lepidoptera, VI.l 

[Philip. Journ. Sci., 19, No. 2. 


Wileman: Japanese Lepidoptera, VI.] 

[Philip. Journ. Sci., 19, No. 2. 



By Warren D. Smith 

Heady Department of Geology , University of Oregon; and Acting Chief, 
Division of Mines, Bureau of Science 



On a recent trip to inspect the mines of the Aroroy district 
of Masbate, a large triangular and forked island just south of 
the southeast end of Luzon, the writer had the opportunity to 
visit some ancient cave-dwellings in the interior of that island 
which was on the whole a most interesting experience. He was 
told of the existence of these caves by Mr. Paul Schwab, an 
American prospector, who has lived in that district since 1905. 
The caves were shown to Mr. Schwab by a Filipino who harvests 
rice in Batwaan Valley each year. It is reported that a 
Mr. Wilson, a lawyer of Masbate, who has some agricultural 
interests in the interior, had visited these caves or similar ones 
and had brought away some relics, but the writer has been unable 
to learn anything about his finds and, as far as is known, noth- 
ing about them has heretofore been published. 

In the preparation of this article the writer has consulted 
Prof. H. 0. Beyer, professor of anthropology in the University 
of the Philippines, but the author alone is responsible for the 
statements herein. No one has as yet made an exhaustive study 
of the caves or the materials found there and the writer wishes 
merely to place on record the information he now has, as a 
preliminary contribution to further study of these and other 
Philippine caves. Similar burial caves have been noted in other 
parts of the Archipelago, but aside from some notes by the 
traveler Jagor,^ with reference to some caves on the coast of 
Samar and Leyte, with Virchow's important appendix, little or 
nothing has been written about prehistoric peoples in the Philip- 
pines, and to the writer's knowledge this is the first description 
of cave dwellings in the Philippines. Many stories and reports 

* Jagor, Fedor, Viajes por Filipinas. Madrid (1875). This is a trans- 
lation from the German. 


234 ^^^ Philippine Journal of Science 1921 

are current about such, but they have not yet got into the 

Mr. E. H. Taylor, formerly of the Bureau of Science, who has 
traveled extensively in the Archipelago making zoological 
studies, has kindly contributed the following memorandum rela- 
tive to some modern cave-dwellers on Coron Island: 

There is a limestone cave on Coron Island, directly across from the 
town of Coron on Busuanga Island, inhabited at the present time by a 
very primitive people known as Tagbanuas. At the time of my visit there 
was no one at home, but there were remains of fire, pots, a hammock, etc., 
testifying to the recent occupation of the cavern. This cave is about 20 
meters above the sea and close to shore. In addition to the articles named 
there was a large kitchen midden consisting mainly of shells of edible 
molluscs and fragments of pottery. Just opposite this site are three small 
limestone islands. In one of these is a very small burial cave containing 
three cadavers, and one very recent coffin made of bamboo and nipa. Ac- 
cording to report in Coron there are many Tagbanua cave dwellers on 
islands north of Palawan and on Culion. 

Geography. — The site of the Batwaan caves is close to the 
trail from Aroroy to Mandaon and about five hours by foot 
from Mandaon and about eight from Aroroy; that is to say, 
about 32 kilometers from Aroroy and almost due south of that 
place (fig. 1). Batwaan valley is about 8 kilometers long 
and 5 kilometers wide, and roughly oval in shape. It is sur- 
rounded by a rim of low hills with fairly steep escarpments 
toward the valley and more gentle slopes away from the valley. 
The caves are found in a large limestone mesa (Plate 1), which 
rises about 100 meters above the floor of a large, flat-bottomed 
valley, through which runs Batwaan Creek. The general 
situation is shown in the sketch, fig. 2. The limestone mesa 
is roughly 0.5 kilometers long and 200 meters wide. The 
formations in this region are generally sedimentary, but this 
valley lies very close to the contact between the igneous rock 
of the central cordillera of the island and the Tertiary sediments 
which dip away from the cordillera to the west. On the trail 
to Batwaan from Aroroy there is much conglomerate exposed, 
and in the hills forming the western rim of Batwaan Valley 
there is conglomerate and sandstone with here and there patches 
of residual limestone. The large limestone mesa in the center 
of the valley is a residual block left by erosion. Formerly 
limestone probably overlay much of this region. This limestone 
is correlated with the lower member of the Malumbang lime- 
stone, which is lower Pliocene or upper Miocene in age. On 
fresh fracture the rock is compact and creamy white; it 

19, 2 

Smith: Ancient Cave Dwellers of Batwaan 


weathers to a blue-gray. The caves in the limestone are the 
result of solution by water. 

The caves. — The caves explored by Mr. Schwab and the writer 
are four in number, but there are several more in the hill. 
These four he has designated as follows: The great living 
cave, the burial cave, the ceremonial cave which is in reality a 
part of the living cave, and the great guano cave. In fig. 2 the 
general situation with reference to these caves is shown. 

Fig. 1. Masbate Island ; outline, showing cave site at X. 

The great living cave is located at the southeast end of the 
hill and has a large opening which lets in plenty of light. The 
cave is somewhat semicircular in shape but is very irregular. 
It is perhaps 15 meters high and 20 meters long, and toward 
the back end of the cave the floor slopes upward to a chimney- 
like opening above. Special features of this and other caves 
of the hill are the large cylindrical holes in the roof. These are 
in some cases about a meter in diameter, but most of them are 
about 8 centimeters in diameter, and they do not go through 


The Philippine Journal of Science 

Fig. 2. Batwaan Valley and caves. 

19,2 Smith: Ancient Cave Dwellers of Batwaan 237 

to the surface; usually swallows and various insects are found 
living in these cylindrical holes. The writer is at a loss to 
explain their origin as there is no indication of water having 
percolated through from the surface and dissolved the material. 
These holes are very symmetrical, very much resembling giant 
drill holes. 

This cave is very dry, and anything found in it might have 
endured through many hundreds of years, as far as decomposi- 
tion is concerned. The floor is somewhat sandy and is made 
up of loose blocks of limestone and fragments of pottery. It 
is quite evident from the amount of material on the surface that 
it would be easy to obtain many kilograms of broken pottery. 
On digging into the floor one comes across bits of charcoal from 
ancient fires, and many marine shells of species of Area and 
Turritella which evidently formed one of the principal articles 
of diet of the people who inhabited the cave. These, being sea 
shells, of course, were brought from the ocean, probably from 
the nearest point, which is known as Nin Bay, about 15 kilo- 
meters to the southwest. There are also many deer teeth and 
broken jawbones of deer, indicating that these people were not 
only fishermen but hunters as well; in fact, this whole cave is 
one great kitchen midden. 

Adjoining the great living cave, and virtually a part of it, is a 
second large chamber in which the floor slopes downward to a 
small outlet on the side of the hill. At the upper end of this 
slope there is a great flat block of limestone which forms a 
natural platform. There is a tradition among the Filipinos 
living in the neighborhood of this hill, one of whom accompanied 
the writer's party in their exploration, that this chamber was 
used in former times for a ceremony connected with marriage. 
There is also a tradition that the bride-to-be had to go into this 
chamber, where she would secure her trousseau from a shelf near 
the top of the cave. The people living in the valley at the pres- 
ent time have never taken part in these rites, but this story has 
been handed down to them with reference to this place. 

The roof of this cave has a great hole open to the sky which 
lights up the whole place and has been instrumental in keeping 
the bats out of this and the living cave. Directly beneath this 
great natural skylight there are some trees growing in the 
center of the cave ; the largest of these is not more than 6 centi- 
meters in diameter and about 10 meters high. These indicate 
that probably the time necessary for them to grow, however 
much that might be, has elapsed since the cave was used. 

238 The Philippine Journal of Science 1921 

The burial cave is a very small one, not over 8 meters from 
the entrace to the back wall, and not large enough to allow a 
person to stand upright in it. It is located on a shelf about 100 
meters to the north of the living cave and perhaps 10 to 20 
meters higher up. There is a very narrow ledge of rock leading 
up to it. Within this cave was found a heap of human bones 
among which there were forty-four human skulls. Here also was 
a head box about 1 meter long which would hold a row of five or 
six skulls. The head box was made of wood resembling molave, 
but was in a bad state of decay. This box had handles carved 
to represent the head of some animal, possibly a crocodile or a 
horse; it was impossible for the writer to tell which. There 
were fragments of a much smaller head box, about 0.5 meter 
in length, with nothing in it. At the far end of the cave there 
was a piece of basket work from which the bottom had rotted 
out. This type of basket work resembles the basket work, as 
far as shape is concerned, now found in the Islands. There was 
also a coconut shell with a hole in the bottom; and, curiously 
enough, in the same cave with these rather recent-looking arti- 
facts were two stone implements. 

On the west side of the hill and somewhat lower down than 
this cave is a great cavern with one fairly large entrance and 
in the rear a chimneylike opening to the surface. This cave is 
nearly semicircular in shape and resembles the interior of a cathe- 
dral. It is quite gloomy and is inhabited by innumerable 
bats. The floor is fairly level and is covered with bat guano. 
Apparently the deposit of guano in this cave is thick. In 
various parts leading off from the central chamber are some 
tunnels leading deep into the interior of the hill ; these were not 
explored. The entrances to these various caves are well con- 
cealed by thick vegetation which has grown up around the lower 
slopes of the hill. The writer regrets exceedingly that the 
views taken of the interior of the caves did not turn out to be 

People inhabiting the cave. — The question as to the kind of 
people who lived here is of course open to some conjecture. It is 
certain that no people living on Masbate Island to-day inhabit 
caves ; nor do the Filipinos living in Batwaan Valley (there are 
only two houses in the valley) know anything about the former 
dwellers in this interesting place, and with the exception of the 
two Filipino men who accompanied the party to the caves the peo- 
ple strictly avoid the locality. The writer has been in caves in 
northern Luzon where the Igorot people have buried their 

19,2 Smith: Ancient Cave Dwellers of Batwaan 239 

dead, but he has. never heretofore found any people living in 
the caves nor found any remains of people formerly living in 
them other than the bones which were buried there. Therefore, 
these caves are particularly interesting because we find the 
artifacts and kitchen middens, indicating not only that people 
lived here but that many people lived in them for a long period of 
time; and, furthermore, we are able to judge pretty well as to 
how they lived. 

The skulls. — Most of the skulls examined showed artificial 
deformation which, history records, was practiced by some of 
the peoples living in the Islands when the Spaniards first came 
here ; but, so far as the writer knows, no people at present living 
in the Philippines indulge in this primitive practice. An exam- 
ination of the skulls (see Plate 2) leads him to believe that an 
earlier race of people than now inhabits Masbate frequented these 
caves. From a preliminary study only. Professor Beyer has 
told the writer that in his opinion the smaller of the two skulls 
was that of a Negrito, the larger, perhaps, of a Chinaman. He 
called particular attention to the rather unusual deformation of 
these skulls. The smaller one, although that of an adult, has 
an open suture along the median line in the anterior part. This 
may have been caused by the use of two separate blocks, one on 
each side, being bound on the head. In the case of the larger 
skull, a totally different method seems to have been used. In 
this there is no flattening, but there is a most unusual depression 
which begins just in front of and above one ear, continuing 
across the top of the cranium down to the corresponding place 
on the other side. It looks as if a band of some metal had been 
bound tightly about the skull in the early years. The frontal 
suture in this case is completely closed. 

The artifacts, — From a study of the artifacts it seems that 
three different cultures are represented, the oldest being repre- 
sented by the stone implements; the second, by the unglazed 
and ornamented pottery ; and the last, by the glazed pottery, the 
badly decayed and rusty fragment of a metal lock, and the basket 
work. The fragments of pottery, shown in Plate 3, are partic- 
ularly interesting, since, so far as is known, nothing like this 
is being made in the Philippines to-day, save among the pagans 
of eastern Mindanao. As a rule, Filipinos do not etch designs 
on their pottery; but, whenever they do attempt decorations of 
any kind (which is the exception) the patterns are painted on 
very crudely. Possibly the patterns on these old fragments from 
Batwaan have Javanese affinities. It seems fairly certain that 
they represent a very early culture stage in the Philippines. 

240 The Philippine Journal of Science 1921 

Plate 4, fig. 1, shows an earthenware vessel of a design with 
which the writer is quite unfamiliar in the Philippines. 

Plate 4, fig. 2, may be either the bottom part of a native stove 
or a fragment of a lid of a giant tinaja (water jar) . 

In Plate 5, fig. 2, is shown a bracelet cut from the top of a 
large Conus shell, while the other two fragments (fig. 1) are 
pieces of pipe bowls. 

We come now to the most interesting of all the artifacts 
found in these caves, namely, the stone implements. So far 
as the writer knows there are only five of these known from 
the Philippines; Mr. Dean C. Worcester has two, the writer 
has one that was given him years ago by a prospector but 
about which he has no data, and the two that are pictured in 
Plate 5, figs. 3 and 4. The one shown in Plate 5, fig. 4, is made 
from a greenish colored felsite, probably a fine-grained diorite, 
and may have been used as a hide scraper. Professor Beyer has 
suggested to the writer that the implement was made from 
this rock because of its general resemblance to jade, from 
which material many Chinese implements were, and perhaps 
still are, manufactured. 

Plate 5, fig. 3, shows an implement of a cherty material, 
which was undoubtedly used as a sort of combination hatchet 
and chisel. These stone implements, some believe, may have 
some historical connection with the ancient Chinese mining 
exploits on this island. The writer is not in accord with that 
view, since tools of this design or of this material would be of 
little use or effectiveness in mining operations. He is of the 
opinion, on the other hand, that these stone implements represent 
a true indigenous stone-age culture in the Philippines belonging 
to the Neolithic Period, 

The writer disclaims any pretense to a special training in 
ethnology; he has described these finds in the hope that 
qualified persons may become interested enough to make fur- 
ther investigations. However, the writer is sufficiently informed 
along those lines, to realize the importance of a complete study 
of this subject, if any safe conclusion as to early movements 
of peoples in the Pacific area is to be arrived at. He agrees 
with a recent statement of Hrdlicka that the solution of Pacific 
anthropology and ethnology will have to be arrived at by way 
of a more complete study of the Continent of Asia and the 
festoons of islands off its east coast. The writer believes that 
the Philippines are in a strategic ethnologic position with re- 
spect to such a study. 


Plate 1 

Limestone mesa, site of the Batwaan caves. 

Plate 2 
Skulls found in the burial cave. 

Plate 3 
Fragments of decorated pottery found in the living cave. 

Plate 4 
Earthenware utensils found in living cave. 

Plate 5 

Fig. 1. Pipe bowls found in the living cave. 

2. A bracelet made of a Conns shell, found in the living cave. 

3. A stone hatchet found in the burial cave. 

4. A stone scraper found in the burial cave. 

TEXT figures 

Fig. 1. Outline map of Masbate Island. X marks the cave site. 
2. Map of Batwaan valley and caves. 

180732 8 241 










Smith: Ancient Cavk Dwellers. 1 

[Philip. Journ. Scl, 19, No. 2. 

Fig. 1. 



^^^^^^^^B^BMBfe-i-t 4bft!iaMMiBBBiMiiy»^ »•■ ^^^^Fvl^H^^^K'/'^tlv. *\ ^1 


^^^^^i^l ':;^|*^||ll 




^^^^■p^ ypp. '.•" ' 




^^^^^^^^^^^^^^^^^^^^^K:1H^^^^^^^^^^^^^^^^^^'''- >■ 


Fig. 2. 

Smith: Ancient Cave Dwellers.] 

[Philip. JouftK. Set., 19. No. 2. 


Smith : Ancient Cave Dwellers.] 

[PIIILIF. JOURN. SCL, 19, No. 2. 


Smith: Ancient Cave Dwellers.] 

[Philip. Journ. Sci., 19, No. 2. 



By Lawrence D. Wharton 

Dean and Professor of Zoology, Junior College of Liberal Arts, University 
of the Philippines, Cebu 


In making postmortem examinations of cats for parasites at 
the College of Medicine and Surgery in Manila, the bile ducts 
of nearly 50 per cent of the specimens examined have been found 
to be infested with a species of Opisthorchis. The first speci- 
mens obtained were thought to be Opisthorchis felineus (Rivolta, 
1884), and I sent a mounted specimen to Dr. H. B. Ward under 
that name. He very kindly sent me drawings of the type speci- 
mens of 0. felineus and 0. pseudo-felinetcs and pointed out 
several differences between my specimen and the other two 
forms mentioned. Since that time I have collected a large series 
of specimens from over twenty cats, all from Manila, and it was 
found that the characters which Doctor Ward had mentioned 
are constant and of sufficient importance to justify considering 
this form a distinct species. I therefore propose the name Opis- 
thorchis wardi for this species. 

Opisthorchis wardi sp. nov. Plate 1, fig. 2, 

Specific diagnosis. — Body elongated, transparent, flat; length 
in preserved specimens, 6 to 9 millimeters; breadth, 1.6 to 2.2; 
living specimens, somewhat larger ; anterior end conical, with a 
slight constriction at the level of the ventral sucker, posterior 
end rounded, occasionally with a small projection around the 
excretory pore; oral sucker 0.17 to 0.28 millimeter in diameter; 
ventral sucker about the same size and about one-fourth of the 
length of the body from the anterior end; pharynx 11^ to 174 
fi in diameter; oesophagus two to three times (as long as the 
pharynx ; intestinal csecse reach almost to the posterior end of the 
body ; excretory pore at the posterior border ; excretory bladder 
narrow, and extending to in front of the testes; testes in pos- 
terior fourth of the body, deeply lobed, anterior with four lobes, 


244 The Philippine Journal of Science 

posterior with five lobes ; ovary in the median line in front of the 
testes, with three distinct lobes, two to the right and one to the 
left; large saclike receptaculum seminis behind the ovary; Lau- 
rier's canal present; coils of the uterus extensive, filling nearly 
all of the middle half of the body; vitellaria lateral, extending 
from behind the ventral sucker to the level of the ovary, acini 
not in distinct groups, no distinct division into anterior and 
posterior portions ; eggs, 28 to 30 ii by 11 ii. 

Habitat and distribution, — Specimens found in the bile ducts 
of domestic cats from Luzon, Philippine Islands. Since coming 
to Cebu to live I have examined many cats with the hope of find- 
ing this parasite, but so far I have not found it on this island. 
I have had no opportunity to look for it anywhere else in the 

The characters which distinguish this species from 0. feli- 
neus (see Plate 1, fig. 1) are the relatively great length of the 
oesophagus, the distinct division of the testes and ovary into 
lobes, the simple arrangement of the vitellaria, and the greater 
extent of the uterine coils. 


Plate 1 
Fig. 1. Opisthorchis felineus (Rivolta), from the type specimen of Doctor 
2. Opisthorchis wardi sp. nov., from the type specimen in collection 
of Doctor Ward; cotypes in collection of L. D. Wharton. 


Wharton: A New Fluke.] 

LPhilip. Journ. Sci., 19, No. 2. 




By S. F. Light 

Professor of Zoology, College of Liberal Arts, University of the 


The Philippine Pennatularia have been little studied, only- 
eleven species being known from the Islands, all from the collec- 
tions of Semper ^ and the Challenger.^ 

The present report deals with the specimens belonging to the 
genus Lituaria in the collection of the department of zoology, 
College of Liberal Arts, University of the Philippines, which 
represent four species, all new to science. 

The only other representative of the suborder Sessiliflorse in 
the collection is the Policella manillensis of KoUiker,^ here re- 
ported for the first time since KoUiker. 

In this report I have adopted the systematic arrangement of 
the group suggested by Kukenthal in his revision,^ and have fol- 
lowed him closely in the diagnosis of the genus Lituaria and the 
key to the species of the genus. 


Polyps arising singly and directly from the rachis. 
Section Pennatulina radiata 

Polyps arising on all sides of the rachis. 

Polyps without a distinct calyx. 

Genus LITUARIA Valenciennes MS., 1850 

Colony club-shaped. Autozooids widely separated, scattered 
or in longitudinal rows. Siphonozooids very numerous, filling all 
the spaces between the autozooids. Axis of varying length and 

^Kolliker, A., Abh. Senckenb. Naturf. Ges. 7 (1869-1870) 109-256; 8 
(1872) 85-275. 

' Kolliker, A., Report of the Voyage of H. M. S. Challenger during the 
years 1873-76 1' (1880). 

•Kukenthal, W., Anthozoa, Pennatularia, Das Tierreich. Franz Eil- 
hard Schulze, 43 Lieferung (1915). 


248 The Philippine Journal of Science 1921 

always four-sided in the rachis region. Spicules are biscuit- 
shaped plates with warts and thorns, crosses, rods, and capstans 
with branched or warted ends. 

This genus was instituted in 1850 by Valenciennes * in his 
manuscript for Pallas's long-known species Pennatula phalloides. 
Recently three other species have been added. Two of these, 
L. habereri Balss ^ and L. hicksoni Thomson and Simpson,^ are 
new. One old species, the Clavella (Sarcobelemnon, Veretilliim) 
australasise of J. E. Gray ^ has been referred to this genus by 
Kukenthal and Brock.^ 

Our collection contains specimens of four distinct types belong- 
ing to this genus, none of which can be identified with any of 
the four known species. These results might well be open to 
doubt were it not for the facts that these are the first species 
of Lituaria reported from the Islands, that the alcyonarian 
fauna of the Philippines is still to a great extent unknown, and 
that where it has been studied it has shown a surprisingly large 
percentage of new species. 

Key to the species of Lituaria, 

1. Axis with longitudinal furrows 2. 

Axis without longitudinal furrows. 

L. australasise (Gray) Kiikenthal and Brock. 

2. Axis with outgrowths on the upper end. 

L. phalloides (Pallas) Valenciennes. 
Axis without processes or outgrowths 3. 

3. Axis not extending to the upper end of the rachis 

L. kiikenthali sp. nov. 

Axis extending to the upper end of the rachis 4. 

4. Rachis of same length as stalk 5. 

Rachis longer than the stalk 6. 

5. Polyp spicules up to 0.12 millimeter long L. habereri Balss. 

Polyp spicules up to 0.20 millimeter long. 

L. hicksoni Thomson and Simpson. 

6. Base of retracted autozooids directed upward, outer spiculated region 

triangular 'L- philippinensis sp. nov. 

Base of retracted autozooids directed outward and upward, outer spicu- 
lated region not triangular 7. 

7. Spicules of rachis averaging 0.13 millimeter in length, with long 

thorns !-• ^^^^^ sp nov. 

Spicules of rachis averaging 0.085 millimeter in length, little branched 
at ends I" ^reve sp. nov. 

* Milne-Edwards, H., and Haime, Palaeontographical Society Monographs, 
Intr. p. 84. 

' Balss, Abh. Bayer. At suppl. 1, p. 81. 

•Thomson and Simpson, Alcyonaria of the Investigator 2: 310. 

' Gray, J. E., Catalogue Sea-Pens, British Museum, 33. 

' Kiikenthal and Br^ock, Ergb. Tief see Exp. 1 3 : 117-170. 

19,2 Light: Notes on Philippine Alcyonaria 249 

Lituaria kiikenthali sp. nov. 

Type. — No. C. 682 in the zoological collection, College of 
Liberal Arts, University of the Philippines; collected in Pore 
Galera Bay on the north coast of Mindoro Island, by 
L. E. Griffin. 

The rather slender, club-shaped colony, from 80 to 120 
millimeters in length, ends in a slender stalk, from 5 to 12 
millimeters in maximum diameter and from one-half to three- 
fourths as long as the rachis. The stalk shows no distinct 
swollen area. The rachis increases in size from the stalk to 
the blunt upper end. At or near its tip it reaches a maximum 
diameter of from 10 to 16 millimeters. The lower part of 
the rachis, while distinctly not a part of the stalk, bears no 
autozooids and in some colonies but few siphonozooids and 
might easily be mistaken for the upper portion of the stalk, 
in which case the stalk would appear to be of about the same 
length as the rachis. 

The axis extends from the midregion of the stalk to a point 
about one-third the length of the rachis from the upper end 
of the colony. It tapers from a blunt end in the rachis region 
to a pointed but not recurved end in the stalk. It is four-sided 
and shows two deep longitudinal grooves in the rachis region 
which join over the blunt upper end. Those portions of the 
colony above and below the ends of the axis are often bent 
at an obtuse angle. 

The large polyps are scattered at fairly regular intervals of 
from 1 to 3 millimeters, but are not in distinct rows. In 
expansion they are from 6 to 8 millimeters in length and from 
2 to 3 millimeters in diameter at the base. They are trans- 
parent with the exception of a triangular brown area on the 
upper surface of the base of each polyp which gives the 
colony, particularly in contraction, a very characteristic spotted 
appearance. This spot fades in alcohol, leaving a transparent 
area. The extensile portion of the polyp contains no spicules 
and is completely retractile within low but quite distinct, 
outwardly and upwardly directed, spiculated basal portions 
which may or may not appear 8-rayed, depending upon the 
amount of contraction of the colony. 

The siphonozooids, which are very numerous, filling all the 
spaces between the autozooids, appear near the base of the rachis 
to lie in crowded but distinct longitudinal rows. 


The Philippine Journal of Science 


The spicules, which are numerous only on the basal, non- 
retractile regions of the autozooids, are very irregular in 
shape and size, having the form of irregular clubs, crosses, 
capstans, etc., usually with divided ends and sculptured surfaces. 
They range in length from about 0.05 to 0.13 millimeter. 
The following are characteristic measurements in millimeters: 

Capstans: 0.117x0.065x0.104; 0.104x0.039x0.052; 0.117 x 
0.0455 X 0.08; 0.0845 x 0.026 x 0.065. 

Clubs: 0.075x0.03. 

Crosses: 0.0975 x 0.039; 0.0845 x 0.039; 0.117 x 0.039. 

Irregular forms : 0.1235 x 0.091 ; 0.104 x 0.039. 

There are no spicules in the inner or outer portions of the 

The rachis is a grayish white, the spiculated regions a dead 
white, and the stalk grayish yellow in formalin or alcohol. 

Measurements of specimens o 

/ Lituaria kiikenthali 

sp, nov. 


Length of — 

Maximum diameter 







1 _ 




































2 — 



5 _ ._. 

6 .__ 



9 . 

10 _. 


12 ___ _ 



15 _ 

16 — 



19 __ 

20 - - . - - 

^ Siphonozooids lacking on lower 15 millimeters of rachis. 
^ Siphonozooids very scarce on lower rachis. 

This species diifers from the four known species of the genus 
in the presence of the brown area on the upper basal portion of 
each autozooid which gives the contracted colony a characteristic 

19,2 Light: Notes on Philippine Alcyonaria 251 

spotted appearance. It differs from all but L. australasise in 
that the axis does not extend to the upper end of the colony, 
and from L. aicstralasise in that the axis is deeply grooved on two 
sides in the region of the rachis and in the large size of its 
autozooids. From L. hicksoni, to which it is apparently most 
nearly related, it differs, aside from the difference in the length 
of the axis and the presence of the brown spot on the upper 
basal portion of the polyp, in the greater length of the rachis 
as compared with the stalk, in that the siphonozooids are ar- 
ranged in distinct longitudinal rows, and in that the autozooids 
are considerably smaller. 

I have named this species after Dr. Willy Kiikenthal, whose 
revision of the Pennatularia has greatly facilitated systematic 
work in that group. 

Lituaria philippinensis sp. nov. 

Type. — No. C. 2459 in the Zoological collection. College of 
Liberal Arts, University of the Philippines ; collected from Port 
Galera Bay, Mindoro, by R. P. Cowles. 

The colony is slender and, with the exception of the lower 
portion of the stalk, rigid. The rachis is somewhat longer 
than the stalk and has a maximum diameter slightly less than 
that of the somewhat swollen lower portion of the stalk. The 
axis, which extends from the midportion of the stalk to the 
extreme tip of the rachis, is four-sided, slender, recurved at its 
lower end, and grooved on two sides throughout its whole length. 
It tapers from the middle of the rachis to the very slender 
lower tip, and less so toward the upper end, which is bluntly 
pointed. The grooves do not join over the upper end. 

The polyps are scattered 'or in indistinct transverse rows or 
whorls, being from 2 to 3 millimeters apart in the long axis of 
the colony, and from 1 to 2 millimeters apart in the transverse 
axis. They are completely retractile within upwardly directed, 
flaplike, basal portions, which have a triangular, spiculated re- 
gion on their outer surfaces. The polyps have transparent 
white walls with black or brown tentacles, or white tentacles 
and a brown stomodseum. The siphonozooids are numerous 
and distinct, irregularly arranged, each with a small nonre- 
tractile basal ^portion similar to that of the autozooids. 

The spicules are unbranched or bluntly branched, sculptured 
capstans averaging 0.13 millimeter in length and 0.025 in 
central diameter. 

252 The Philippine Journal of Science 

Measurements of specimens of Lituaria philippinensis sp, nov. 


Length of — 

diameter of— 






1 .. 








2 .. ._ 


This species differs from L. phalloides in the absence of any 
outgrowth from the upper end of the axis ; from L. hicksoni in 
that the colony is much slenderer, that the nonretractile basal por- 
tions of the autozooids are of quite different shape and appear- 
ance, and that the spicules are smaller; from L. aitstralasise in 
that the axis is deeply grooved; and from L. kukenthali in that 
the axis extends to the upper end of the rachis and in the 
absence of the characteristic brown area on the upper part 
of the polyp base. 

From L. habereri, to which it is apparently most nearly re- 
lated, it differs, so far as I have been able to determine from 
the rather meager description in Kukenthars revision (I do 
not have access to Balss' description), in that the rachis is 
longer than the stalk; that the axis does not extend through- 
out the entire colony, ending below in the middle of the stalk; 
in that there are no spicules, in the stalk rind ; and in that the 
colors are different. 

Lituaria moUe sp. nov. 

Type. — No. C. 2457 in the zoological collection. College of 
Liberal Arts, University of the Philippines ; collected from Port 
Galera Bay, Mindoro, by L. E. Griffin. 

The rachis, which is somewhat longer than the slender stalk, 
increases in size from the stalk to the blunt upper end where 
it reaches a maximum diameter of 10 millimeters. The entire 
colony is expanded and soft : hence the specific name. 

The large, irregularly scattered autozooids reach a length of 
10 millimeters and a basal diameter of 2 to 3 millimeters. The 
upper portion of the autozooids and the tentacles are brown. The 
numerous long, closely set pinnules are transparent white, as 
are the polyp walls. The lower portion of the stomodseum is 

19,2 Light: Notes on Philippine Alcyonaria 253 

brownish black and shows through the transparent polyp walls. 
The siphonozooids are few and scattered. 

The four-sided axis extends from about the midregion of the 
stalk to the tip of the rachis. It is slender, pointed, and reflexed 
below and increases in size to a point near the upper end from 
which it tapers slightly to the bluntly pointed tip, showing two 
deep longitudinal furrows which do not meet over the upper jend. 
There are no polyp spicules. Those of the rind of the rachis 
are rather large capstans and stars with much-branched ends. 
They average 0.11 millimeter in length and are scattered, being 
thickest in the region around the base of the polyp, form- 
ing, in the upper region of the rachis, fairly distinct verruca- 
like structures. 

The stalk rind contains scattered, unbranched, sculptured 
clubs, plates, and capstans. The colony, aside from the polyps, 
is yellow in formalin. 

Measurements of a specimen of Lituaria molle sp. nov, 

Length of colony 130 

Length of rachis 70 

Length of stalk 60 

Maximum diameter of rachis 10 

Maximum diameter of stalk 9 

This species, while based on a single specimen, seems distinct 
enough to stand without confusion. It differs from the other 
three Philippine species, among other things, in having spicules 
in the stalk rind; from L. aicstralasise in having a grooved 
axis; from L. phalloides in having no outgrowths from the 
axis ; from L. habereri and L- hicksoni in having a rachis longer 
than the stalk and in details of spiculation and in color. 

Lituaria breve sp. no v. 

Type. — No. 2458 ^ in the zoological collection. College of Liberal 
Arts, University of the Philippines; collected from Port Galera 
Bay, Mindoro, by S. F. Light. 

The rachis is about twice as long as the slender sfalk. The 
colony is slightly curved and tapirs from about the middle of 
the rachis to the two ends. The autozooids are few, large, and 
irregularly scattered. In contraction they lie in outwardly and 

• This species, like the others described in this paper, was described some 
time ago. Since then, unfortunately, the type specimens have been mis- 
placed. In view of the very distinct characters of the species, however, 
I have considered it permissible to publish the description, in spite of the 
loss, temporary it is to be hoped, of the type specimens. 

254 ^^^ Philippine Journal of Science 1921 

somewhat upwardly directed, heavily spiculated, verruca-like 
basal regions about 2 millimeters in diameter, whose large size 
in proportion to the diameter of the rachis gives the colony a 
characteristic, irregular appearance. 

The siphonozooids are numerous and irregularly arranged 
and have spiculated nonretractile basal regions, similar to those 
of the autozooids, and are about 0.8 millimeter in diameter. In 
the lower one-third of the rachis the zooids are nonspiculated and 
8-rayed, brown on the outer surface. 

The axis is cylindrical in the stalk region ending in a 
slender recurved tip. It is four-sided in the rachis region, 
shows two shallow longitudinal furrows, and tapers toward 
the tip where, for a distance of about 10 millimeters, it is 
slightly roughened. 

There are no spicules in the retractile portions of the 
polyps, in the stalk rind or in the rind of the lower part of 
the rachis. Those in the upper portion of the rachis are present 
only in the verruca-like basal portions of the autozooids and 
siphonozooids. They are small, averaging about 0.085 milli- 
meter in length, sculptured, but little-branched, consisting mainly 
of plates constricted in the middle, crosses, and capstans. 

The two colonies differ decidedly in detail due to the fact that 
one is immature and the larger colony has evidently been 
broken at some time and r^igenerated. The above description is 
an attempt to give a specific diagnosis. A study of new 
specimens will probably lead to some changes, but the species 
is so clearly distinct that it seemed justifiable to establish 
it here. Below are separate descriptions of the two speci- 
mens, in so far as they differ. 

Specimen A. — Apparently a mature specimen. Length of 
colony, 69 millimeters; of rachis 49; of stalk 20. Maximum 
diameter of rachis 6 millimeters, stalk 4. Stalk constricted at 
point of union with the rachis. Rachis tapering to a point. 

The axis, which extends from near the base of the stalk to 
the tip of the rachis, consists of two parts; the upper part, 
25 millimeters long, has beccfme attached throughout most of 
its length to the flat ungrooved side of the lower portion. 

Specimen B; immature. — A very slender colony 40 milli- 
meters in length. Rachis and stalk of about the same length, 
due apparently to the fact that the 8-rayed zooids found in 
the lower portion of the rachis of the mature colony have not 

19,2 Light: Notes on Philippine Alcyonaria 255 

developed. The autozooids and siphonozooids are somewhat 
smaller than in the larger colony, but they have the same 
form and arrangement. The spiculation is the same. 

The similarity in the form of the colony, in the arrange- 
ment and form of the autozooids and siphonozooids, and in the 
spiculation mark these two specimens as belonging to the 
same, very characteristic new species. 


^ \y&. 

Vol. 19, No. 3 

September, 1921 

The Philippine 
Journal of Science 




Published by the Bureau of Science of the Government of the Philippine Islands 

Elhsr D. Merrill, M.S.» Editor 

R. C. McGregor, A.B., Associate Editor 

Albert H. Wells, A.B.; Granvillb A. Perkins, Ph.D.; A. P. West, Ph.D. 
T. Dar Juan, A.B., Phar.D.; F. Agcaoili, A.B.; A. S. Arguelles. B.S. 

Albert E. W. Kino 


Warren D. Smith, Ph.D.; Roy E. Dickerson, Ph.D. 


H. W. Wade, M.D.; Otto Sch5bl, M.D. 

P. G. Haughwout; Stanton Youngberg, D.V.M. 

Experimental Medicine 

Liborio Gomez, M.D., Ph.D.; P. Caldbron, B.A., L.M. 

Vicente de Jesus, M.D. 

Clinical Medicine 

W. H. Brown, Ph.D.; C. F. Baker, M.A.; Atherton Leb, M.S. 
r 0. A. Reinking, B.S.A., M.S.; L. M. Guerrero, Phas.D. 


Albert C. Herrb, Ph.D.; C. F. Baker, M.A.; S. F. Light, M.A. 

C. S. Banks, M.A.; L. D. Wharton, M.A.; W. Schultzb 


H. O. Beyer, M.A.; Otto Johns Scheerer, M.A. 

Anna B. Banyba, Copy Editor 

Manuscript intended for publication and books for review should be sent 
to the editor. One hundred separates of papers published in the Journal 
are furnished to authors without charge. Additional copies may be had at 
authors' expense. 

Publications sent in exchange for the Philippine Journal of Science 
should be addressed: Library, Bureau of Science, Manila, P. I. 

The Journal is issued twelve times a year. The subscription price is 6 
dollars, United States currency, per year. Single numbers, 50 cents each. 

Subscriptions may be sent to the Business Manager, Philippine Jour- 
nal of Science, Bureau of Science, Manila, P. I., or to any of the agents 
listed on the third page of this cover. 

The Philippine 
Journal of Science 

VOL. 19 SEPTEMBER, 1921 No. 3 


By J. Preston Maxwell' 
Of the Peking Union Medical College, Peking, China 


Among the causes of chronic invalidism and economic loss 
in China, filariasis may well be considered worthy of a place. 
It is true that the effects are not so striking in its influence 
on the death rate as those of plague among acute affections, or 
malarial fever and dysentery among endemic diseases ; neverthe- 
less in certain regions it plays a large part in diminishing the 
working capacity of a considerable number of the manual work- 
ers and in rendering many of them altogether incapable of 
work. The aetiology of the disease, considered in its broad 
aspect, is simple. With the one exception of an ocular filaria 
described first in China by Stuckey and Houghton (43) and 
now classified by Leiper(i8) among Thelazia, there is only one 
filaria in China to be found in man; this is Filaria bancrofti, 
Filaria perstans has been seen once, but not in a Chinese, and 
there was no doubt that the infection had been acquired on the 
west coast of Africa, where the patient had previously resided. 

Thus the subject of filariasis resolves itself into infection 
with Filaria bancrofti. This infection is carried out in China 
as elsewhere by the agency of the mosquito. Culex fatigans 
and C. pipiens are both common in China, although it is possible 
that there are other varieties of the Culicidse that are potential 
carriers of the disease. According to Button (14) one variety 
of Anopheles is also a potential carrier. Post-mortem examina- 

* Before coming to Peking, Dr. J. P. Maxwell was engaged in medical 
mission work in Fukien Province; this paper represents the results of 
twenty years* study of the subject and includes material that was compiled 
for a thesis, but which has never been published. 

181052 257 

258 ^^^ Philippine Journal of Science 1921 

tions of its victims are still few in number, and there is much 
yet to be worked out concerning its morbid pathology. 

We do not yet know, for example, how long it takes after 
infection before the embryos appear in the blood stream ; nor do 
we know why hyperfilariasis does not take place in those who are 
continuously exposed to the bites of infected mosquitoes. We 
do not know how long the adult worms are capable of putting 
forth embryos, nor do we know the true cause of filarial 
periodicity. Occasionally it is possible for the parent worms 
to die as the result of a catastrophe, which may involve the 
human host and cause his severe illness or even death; and if 
this has taken place and the human host has survived, it is 
possible for the latter to become reinfected and again present 
the phenomena of the disease. 

Yet again it is possible for the human host to harbor the 
parasite and present no symptoms of disease at all, the only 
evidence being the presence of the embryos in the circulating 
blood at the proper time. 

The distribution of the disease throughout China is somewhat 
peculiar. Roughly speaking, the infection does not spread north 
of the Yangtse Valley, though individual cases of the disease may 
be met with farther north. These individuals, however, have 
been infected in the southern regions and not in the northern. 

Following the line of the Yangtse River west, the disease is 
found sporadically along both banks and also in Kweichow and 
southern Szechwan, getting less and less frequent as one ascends 
the river. 

Coming back to the coast from the mouth of the Yangtse 
down to the Tonquin border there is a belt some 15 to 25 miles 
(25 to 40 kilometers) broad, and the major portion of the 
disease is found in this coast belt. Most of the islands off the 
coast are also infected, but not heavily. When one passes in- 
land beyond this belt, the infection is practically lost, although 
occasionally a small patch of infection may be found on the 
higher reaches of a river; for it tends to spread upward along 
th£ banks of all the rivers between the mouth of the Yangtse 
aM the Tonquin border, but not to any considerable distance. 
Kiangsi is entirely free, and Fukien away from the coast belt 
is uninfected, and when imported inland, except in the larger 
river vicinities, the disease does not, according to my experience, 
tend to spread. 

The case incidence, of course, varies with the region and, speak- 
ing broadly, increases toward the south and decreases toward 

19.3 Maxwell: Filariasis in China 259 

the north. Taking a large hospital about the middle of the coast 
belt, I found a percentage of 2.4 of filarial cases, who came to 
the hospital for some disease connected with this infection 
(Changpu, Fukien) ; but this does not represent the incidence 
of filarial infection in this region. My figures give a percentage 
of 24.8 of the general population infected with the parasite, 
and the work of my successor at that hospital. Dr. J. H. Mont- 
gomery, shows a slightly higher figure. 

Besides these figures, 3.39 per cent of the general population 
showed no embryos in the blood, but presented signs of old 
filarial disease, while 16.1 per cent of another series were found, 
on microscopical examination, to be infected but presented no 
signs and had no history that could be attributed to the presence 
of filaria. 

With regard to the age incidence, taking a series of 67 cases 
affected with filarial disease, we have the following findings : 

Table 1. — Showing age at which disease commenced in sixty^eight cases. 

Ages in years. C^geg^ 

1 to 10 1 

10 to 20 e 

20 to 30 22 

30 to 40 25 

40 to 50 10 

50 to 60 3 

Above 60 1 

Table 2. — Showing duration of disease before patienVs first visit to 


Years. Cases. 

1 7 

1 to 5 28 

6 to 10 17 

11 to 15 9 
15 to 20 6 

According to the nature of their disease, a series of two hun- 
dred sixty hospital patients can be classified as follows : 

Table 3. — Two hundred sixty patients, classified according to disease. 


Elephantiasis of scrotum 48 

Lymph scrotum 44 

Elephantiasis of right leg 35 

Elephantiasis of left leg 43 

Elephantiasis of both legs 13 

Filarial abscess ^ 33 

Filarial gangrene of scrotum 8 

Lymphatic fistula q 

Chyluria 4 

Other filarial diseases 26 

260 The Philippine Journal of Science 1921 

All classes and varieties of individuals may become subjects 
of filariasis. The oldest patient in this series was 64 years 
of age; the youngest patient was 13, but patients may be in- 
fected at a still younger age. 

There is no doubt that for some reason the female sex in 
this region is not so subject to filariasis as is the male sex. 
This may be partially due to dress, the women keeping their 
legs and ankles much more covered than do the men. 

Of the two hundred sixty cases listed in Table 3, only five 
were women. 

I am unable to give the statistics of infection among women 
in the general population, as they are so superstitious that there 
has been great diflficulty in obtaining specimens of their blood; 
but granting that the mosquito-infection theory is correct, there 
is no absolute reason why anybody and everybody should not 
become infected. 

There is no doubt that the field laborer is more subject to 
this disease, and to its severer forms, than the literary man and 
the shopkeeper. Probably the amount of leg that is normally 
bare contributes to this fact, and the rough, dirty work greatly 
assists such diseases as elephantiasis of the scrotum and leg. 

It has been surmised that the fisherman is especially sub- 
ject to the disease, but there is not sufficient evidence to support 
this assertion. Very often the immediate inhabitants of the 
seaboard are badly off and do not get sufficient to eat, and 
the exposure to salt-laden wind and blowing sand would assist 
the development of elephantoid changes. 

Let us now turn to the parasite itself and the diseases of 
which it is the cause. The embryonic form of Filaria ban- 
crofti was first discovered in 1863 by Demarquay(i3) in a case 
of chylous dropsy of the tunica vaginalis. In 1866 Wiiche- 
rer(46) found the same form in a case of chyluria. In 1870 
Lewis (19) confirmed this in a case in India. In 1872 the same 
observer discovered that the blood of man was the normal 
habitat of the parasite. In 1874 Sonsino,(42) without knowl- 
edge of Lewis's discovery, made the same observation in Egypt. 
The parental form was first discovered by Bancroft, sr.,(2) 
in Australia. 

The discovery of the mosquito as the intermediate host was 
made by Manson in 1877; (24) and Bancroft, jr., (3) in 1899, 
demonstrated that the complete metamorphosis takes about six- 
teen days. 

Finally Low, (20) in 1900, showed that the filaria, after at- 
taining its proper development, makes its way into the proboscis 

19,3 Maxwell: Filariasis in China 261 

of the mosquito, apparently with a view of leaving the mosquito 
when it bites, and this migration generally takes place in pairs. 

The parent filarise ^ have been found many times. They are 
nematode worms, both sexes being found in the human body 
''often inextricably coiled about one another.'' They are some 
3 to 4 inches (7.6 to 10.2 centimeters) in length, hairlike, and 
transparent. They have been found in many situations: in 
lymphatic trunks, in lymphatic varices, in varicose lymph glands, 
in the tissues removed in operating on lymph scrotum, in the 
tissues removed in operating on elephantiasis of the scrotum, 
and in filarial abscesses. The female worm is the larger of 
the two. In length, it is from 3 to 4 inches (7.6 to 10.2 cen- 
timeters) ; in breadth, about 1/90 of an inch (0.2 millimeter) ; 
the greater part of the body is occupied by the two uterine 
tubes, containing ova in all stages of development. The head 
is club-shaped and simple, and the tail is tapered and rounded 
off, with the anus opening just in front of its termination. The 
vagina opens near the mouth. The cuticle is smooth and devoid 
of markings. 

The male worm is very slender and has a marked tendency 
to curl. The extreme end of the tail is sharply incurved. The 
cloaca gives exit to two unequal, slender spicules. Caudal pa- 
pillae are present. These parent filarise have considerable mo- 
bility and can be kept alive in salt solution for some hours. 
Having described the parental forms of the worm, let us look 
at the characters of their progeny. 

The ova are not normally found in the blood or lymphatic 
system; in fact, the appearance of these probably means that 
parturition has not gone on normally and has an important 
bearing on the question of lymphatic obstruction. I have met 
with them once in a case of lymph scrotum, in a microscopic 
preparation got by tapping the lymphatic system of that part, 
and in this case repeated examination failed to find the embryo 
in the blood. On the other hand, the embryo could be readily 
made out coiled up inside the ovum. The size of these ova 
is stated by Manson(25) to be about 1/500 by 1/750 of an inch 
(0.051 by 0.034 millimeter). The embryo coiled up in the ovum 
gradually stretches its chorional envelope, which probably forms 
the sheath of the mature filarial embryo. 

In describing the embryos of Filaria bancrofti, as commonly 
met with in the blood, I largely follow Hanson's description, 

^ For a full description of the parent filariae and their embryos see 
Tropical Diseases. Cassell, London (1917) 681-702. 

262 The Philippine Journal of Science 1921 

having verified the same in all points. When examined in a 
fresh specimen, the worm is easily made out as a transparent, 
colorless, snakelike organism, which wriggles about very actively 
among the blood corpuscles, but which does not rapidly pass 
out of the field of vision. It has a long, slender body, inclosed 
in a delicate, structureless sheath, in which it easily moves 
backward and forward. This body is blunt at the one end and 
pointed at the other. It is about 1/80 of an inch (0.317 mil- 
limeter) in length and 1/3000 of an inch (0.0085 millimeter) 
in breadth. It is most easily found with a 1-inch objective, but 
for the details a higher power must be employed. 

Carefully examined under a higher power and with the as- 
sistance of a staining reagent, a musculocutaneous layer is made 
out, in which delicate transverse striation can be discerned. 
An indefinite viscus can also be made out in the center portion 
of the worm, rather toward its posterior end. A shining V- 
shaped spot is clearly to be seen about the junction of the head 
fifth with the remaining four-fifths of the body. This spot is 
well shown on staining with dilute hsematoxylin, but is readily 
seen in unstained specimiens. By staining, however, a second 
spot is also brought out a short distance from the tail, much like 
the first in form. The exact meaning of these spots is not 
quite clear, but it is suggested that they are connected with 
development. In the interior of the musculocutaneous cylinder 
is a mass of cells whose nuclei are brought out well by staining. 
There is a break in this central column of nuclei at a spot just 
posterior to the anterior V-spot, but the significance of this 
break is not yet understood. 

The head is covered by a delicate 6-lipped prepuce, and a short 
fang is occasionally to be made out, protruding from the head, 
in specimens where the worm has almost ceased movement. 

One of the most notable things about these filarise is the 
feature that goes by the name of filarial periodicity. In the 
case of the embryos of Filaria bancrofti it is the rarest thing 
to find one of these in the peripheral blood during the day. 
About 5 or 6 o'clock in the afternoon a few begin to appear, and 
the number gradually increases until the maximum is reached, 
about midnight. The number then gradually diminishes until 
about 6 or 7 o'clock in the morning, when they disappear from 
the blood altogether till the following evening. If, however, 
the filarial subject be made to sleep during the day and rise at 
night, it is possible to reverse these conditions and cause the 
filarise to appear during the day instead of at night. In the 

19,3 Maxwell: Filar lasis in China 263 

Philippines it is known that this periodicity is usually lacking, 
and by the acetic acid concentration method, even where the 
regular periodicity prevails, it is possible to show that microfi- 
lariae are never entirely absent from the peripheral blood. 

What is the cause of this periodicity? To answer that it is 
a provision to enable them to be reached by their intermediate 
host is only to throw the question a step further back. What 
then is the cause that leads them to forsake the periphery during 
the day? Possibly it has to do with the accumulation in the 
body of some chemical substance which, while producing sleep, 
tends to attract the filarise to the periphery. In this connec- 
tion an article by Lynch ^ is of importance as . confirming the 
views of Smith and Rivas that the mechanics of the capillary 
circulation play a large part in the production of filarial perio- 
dicity. He gives details of some experiments with drugs which 
alter the vessel tone. A vasodilator such as nitroglycerine was 
followed by a decrease of the embryos in the peripheral blood, 
while vasoconstrictors such as epinephrin or pituitrin were fol- 
lowed by an increase of these embryos, and a collapsed lung 
in a dog accumulated microfilariae immitis in enormous num- 

Their absence from the liver and spleen during the day, 
as proved by the results obtained from aspirating these organs, 
is remarkable. These results have been completely confirmed 
by the post-mortem examination reported in 1899, by Man- 
son, (26) of the case of a man harboring this parasite who 
committed suicide during the day by swallowing prussic acid. 
In this case the liver and spleen were both practically free 
from embryos, and both of these organs are markedly con- 
cerned with metabolic processes. 

Where do the embryos retire to during the day? In the case 
of suicide above quoted they were found to be distributed as 
follows : 
Table 4. — Distribution of embryos in body of man dying in the daytime. 

Lung Many. 

Large vessels near heart Many. 

Vessels of heart wall Many. 

Brachial venae comites Moderate number. 

Liver Almost absent. 

Spleen Almost absent. 

Brain A few. 

Scrotum None. 

' Lynch, K. M., Filarial periodicity, Journ. Am. Med. Assoc. 73 (1919) 

264 The Philippine Journal of Science 1921 

How they, in their sheaths, manage to maintain their posi- 
tion in the larger vessels against the blood stream is as yet 

So far we have dealt with the parasite in its human host; 
now we must deal with its progress through its intermediate 

By artificial means it is quite possible to make the filarial 
embryo cast its sheath. One of the best ways is to lay the 
preparation in an ice box overnight. When the slides warm 
up next day, wherever the blood has become laky, the filarise 
endeavor to break through their sheaths, and in a short time 
they succeed in doing so and move very freely about the slide. 

After the blood has passed into the mosquito's stomach, the 
same proceeding takes place. A few hours after feeding, this 
can be easily observed, and in another few hours the mosquito's 
stomach is found to contain only the empty sheaths. Where 
have the filariae gone? As soon as they lose their sheaths, 
they become able not only to move about freely, but also to pierce 
the stomach wall. 

They are found to have passed into the muscles of the mos- 
quito's thorax. There they undergo a metamorphosis lasting 
some sixteen to twenty days, a proceeding which has for its re- 
sult the formation of a mouth, an alimentary canal, a peculiar, 
trilobed tail, and finally a parasite much grown in size and 
activity. The metamorphosis being completed, they pass for- 
ward for the most part to the cephalic region, a few only 
passing backward into the tissues of the abdomen. 

From the head they pass often in pairs into the proboscis, 
and are conveyed thence into the human host when the 
mosquito bites. They lie in the proboscis between the under 
surface of the hypopharynx and the upper surface of the labium, 
extended in the loose connective tissue of the latter. 

As far as my researches go, in at least 95 per cent of the 
cases especially examined for eliciting the source of infection 
there was either a near relative, or a neighbor, or some one of 
the same village to be found as a possible source. But there 
were also cases like that shown in Table 5. 

In this case the mother had lived with the father for at 
least twenty years after he started elephantoid fever, but she 
had never suffered from any form of filarial disease. Her blood 
was quite free from the parasite. The sons, who were infected 
and noninfected, had always lived together and slept two in one 
bed and three in another. The two infected ones had acquired 


Maxwell: Filariasis in China 


the disease about thirteen and twelve years previously; at least 
their attacks of fever began about that time. Although they 
had continued to live cooped up in this way, no other member 
of the family had been infected. 

Table 5. — Showing irregular incidence of infection of one family by filariss. 

Age. Condition. 

Mother _ 







Dead; had slight elephantia- 
sis of leg. 


Bought. No sign of disease. 




Three years ago one of the infected individuals went to 
live in a separate house with his aunt and cousin. Although 
mosquitoes infected with filarise have been found by me in this 
house, and there is nothing to prevent the members of the 
household being freely bitten, yet this aunt, aged 50, and cousin, 
aged 13, remain uninfected. The other infected individual 
married two years ago, and his blood literally swarms with 
filarise, yet his wife at the present time shows no sign of disease. 

How are these facts to be explained? It may be declared 
that in order to bring about a successful infection the mosquito 
bite must be in a region richly supplied with lymphatics and 
that probably this has not been the case in these persons. Both 
of the infected brothers are, otherwise, strong and healthy, and 
they were in good health when the attacks began. This tends 
to negative any theory that rests on the premise of lowered 
vitality as the cause that makes siiccessful infection take place. 

The probability that a single individual can be infected more 
than once, and that many mature worms may be found in the 
same subject, seems on the other hand to indicate that, in some 
instances at least, infection is not a work of great difficulty. 

It is very probable that many people are bitten by the fllaria- 
infected mosquito, with the result that the worm passes into 
their tissues, but it does not follow that embryos appear in the 
blood of such patients. For the appearance of these, it is 
necessary that the female worm should be present and should 
have been fertilized by the male. 

Let us suppose that the male fails to find the female. 
Possibly they die and are absorbed without giving trouble. 

266 ^^^ Philippine Journal of Science 1921 

At the same time, it is possible that a single worm might, by- 
setting up an inflammatory process, lead to the blockage of a 
main lymphatic trunk, in which event embryos would be absent, 
and yet the case might present some of the characters of filarial 
disease. It is quite possible also that such a worm in its death 
might give rise to filarial abscess. 

In this connection one may perhaps refer again to the ten- 
dency of the worms to pass in pairs into the proboscis of the 
mosquito ; it may be intended by nature thus to provide, by the 
introduction of a male and a female, for reproduction. 

The whole subject is one that needs careful working out, in 
view of its importance in devising prophylactic measures. For 
the present no doubt the mosquito net is the reasonable means of 
prophylaxis, each infected individual being a source of danger 
to the community. 


For examination of fresh specimens, it is only necessary to take 
a large drop of the patient's blood, cover it gently with a 
cover glass, and examine it under a low power of the micro- 
scope, when the filaria is easily distinguished by its movement. 
If such preparations are ringed with vaseline, it is possible 
to preserve them for some time: at any rate, the filarise may 
be kept alive for at least a week. 

For systematic study it is better to keep to a fixed time. In 
these researches for statistical purposes all the specimens were 
taken between 9.30 and 10 o'clock in the evening. As my pa- 
tients used to go to sleep at an early hour, and were up at 
daylight, this time was late enough. 

For this class of work it is better to use specimens prepared 
in the following way: The blood is taken from the finger or 
ear, a lancet or triangular needle being the best instrument for 
the purpose. Several drops of blood should h6 received on a 
glass slip and spread out gently with the needle. The slide 
must then be laid aside on a flat surface and covered, to prevent 
dust spoiling the specimen. Two specimens at least should be 
taken from each patient. As soon as convenient, and the sooner 
the better, these slides, which must be already dry, are fixed 
by warming them over a spirit flame, and then are immersed 
in water. This dissolves out the haemoglobin and leaves a 
colorless specimen, which is then placed in a watery solution 
of methylene blue for a few minutes, passed through water, 
and examined wet without a cover glass under a 1-inch ob- 
jective. The filariae are easily distinguished, being stained a 

19,3 Maxwell: Filariasis in China 267 

distinct blue. The white corpuscles take the dye; the red 
corpuscles remain unstained.* In this work a mechanical stage 
with parallel movement is quite indispensable. If it is wished 
to prepare permanent specimens of filarise, a patient should be 
selected in whose blood there are many embryos. A very thin 
preparation of blood should be taken and fixed with alcohol 
or heat. The haemoglobin is washed out by water acidulated 
with a little acetic acid; the specimens are stained with any 
stain that may be wished and are mounted permanently in 
balsam. For single stains, hsematoxylin or methylene blue are 
very good, while for double staining hsematoxylin and eosin 
make a good combination. 

I have already stated that the staining should be done as 
soon as possible. It is possible to leave the slides in a dry 
place for months and yet get a satisfactory result, but the 
removal of the haemoglobin becomes more difficult, and the 
specimen does not stain well. It need hardly be said that, 
in taking the blood, due care must be taken. The needle or 
lancet must be carefully sterilized, and the skin of the part from 
which the blood is to be drawn should be cleaned first with soap 
and water and then with spirit. If this is done, there are never 
any unpleasant effects following. 

The acetic acid concentration method of Smith and Rivas ^ 
is of great use in giving a quantitative estimation of micro- 
filariae in the blood. 

The method is as follows : 

From 0.1 cc. to 1.0 cc. of blood is tjaken from the finger and collected 
in 5 cc. of a 2 per cent acetic acid solution for the purpose of laking the 
blood. The mixture is shaken gently for several minutes and then centri- 
fuged and spreads are made from the sediment. 

Lynch ^ used 1 cubic centimeter of blood in 10 cubic centi- 
meters of 2 per cent acetic acid solution, centrifuged, washed 
and recentrifuged several times, spread the sediment on a slide 
and counted the whole number in the sediment. 

During 1902 Gulland(i5) suggested the possibility of the 
diagnosis of filariasis from a differential count of the leucocytes 
in the blood of the patient. He pointed out that filariasis is 
accompanied by a leucocytosis proportional to the number of 

* For this method, which works very well, I am indebted to Sir Patrick 

° Smith, A. J., and Rivas, D., Notes upon human filariasis. Am. Joum. 
Trop. Dis. and Preventive Med. 2 (1914) 368. 

° Lynch, K. M., Filarial periodicity, Journ. Am. Med. Assoc. 73 (1919) 


The Philippine Journal of Science 


filarise found in the peripheral blood. Not only are the leuco- 
cytes increased in number, but there is a distinct condition of 
eosinophilia, which also varies in proportion to the number of 
filariae present. Gulland's observations have been confirmed by 
Coles, (10) whose percentages are given in Table 6. 

Table 6. — Cole's differential count percentages. 




Largre uninucleated 

Case I. 

Per cent. 





Case II. 

Per cent. 



Per cent. 




2, not above 3. 

There is also a constant high proportion of lymphocytes, 
while there is a low percentage of polymorphonuclears. 

At first sight it looks as though the data given in Table 6 
might be a method helpful to diagnosis, but so far as I can 
judge the range of usefulness is very small; for, in the first 
place, it is just in the cases where the embryos are very diflficult 
to find that the eosinophilia is least and therefore inconclusive ; 
and, in the second place, Ascaris, Oxijuris, Tasnia, and Ancylos- 
toma are all prone to produce an eosinophilia. 

In the region in which I worked the majority of the people 
were infected with Ascaris, while ancylostomiasis and other 
worm infections are by no means rare. With this knowledge, 
given a doubtful patient with a slight degree of eosinophilia, 
one would hesitate to say, on this ground alone, that he was 
the subject of filar iasis; while, if the parasite be found in the 
blood, the method is unnecessary for diagnostic purposes. 

A very valuable paper on this subject has been written by 
Whyte of Swatow, China. (44) Calvert (6) argued that eosino- 
philia is greatest when microfilariae are absent from the peri- 
pheral blood; but Whyte hesitates to accept this view, and I 
agree with him. 


In the first place, it must be again borne in mind that often 
the filarial worm gives rise to no inconvenience whatever and 
appears to be absolutely innocuous. It may be harbored for 
years without the host having any idea of its presence. The 
diseases dependent on its presence may be classified as follows : 

19,3 Maxwell: Filar iasis in China 269 

I. Inflammatory mischief, either directly or indirectly connected with 
the parasite. 

1. Elephantoid fever. 

2. Lymphangitis. 

3. Erysipelatoid inflammation. 

4. Dermatitis and cellulitis. 

5. Abscess. 

6. Orchitis. 

7. Acute arthritis or synovitis. 

8. Gangrene of the scrotum. 

9. Filarial haemoptysis. 

II. Disease due to obstructive interference with the lymphatic system. 

1. Lymphatic varix. 

2. Lymphatic fistula. 

3. Varicose groin glands. 

4. Lymph scrotum. 

5. Chyluria. 

6. Chylous dropsy of the peritoneum. 

7. Chylous dropsy of the tunica vaginalis. 

8. Chylous diarrhoea. 

9. Elephantiasis scroti. 

10. Elephantiasis vulvae. 

11. Elephantiasis of the legs. 

12. Elephantiasis of the arms. 

13. Elephantiasis of the mammae. 

14. Elephantiasis of limited skin areas. 



Elephantoid fever is somewhat of a misnomer, as the fever 
is by no means always followed by elephantiasis ; but its meaning 
is now so well known that there is no fear of confusion. 

Fig. 1 shows very well the characteristics of a typical case. 
A filarial subject may be attacked in one of two ways. In the 
one form, after an hour or two of malaise, he is suddenly taken 
with a rigor, which may be exceedingly severe, but which is 
generally of moderate severity. The temperature rapidly rises, 
and the patient feels very ill, with a hot and burning skin and 
sometimes marked nausea. This condition persists for from 
forty-eight to seventy-two hours, and then the temperature falls 
with moderate rapidity, reaching the normal in from twelve 
to twenty-four hours. This fall is often accompanied by profuse 
sweating. In the other form there is no rigor, but there is 
malaise for some hours before the attack. There is also marked 
uneasiness or tenderness in some lymphatic region. The tern- 


The Philippine Journal of Science 


perature rises rapidly, and the patient is as ill as in the preceding 
case, and the course and the termination of the affection are 
much the same. 

It must be clearly understood that elephantoid fever may 
take place 'without any external signs of lymphangitis; on the 
other hand, very often there are definite signs of inflammation 

V."^cD 'tr^ *^ "12 'cm '- "O ^OJ-^ "J. 
OOOOOoOO O:),^! Oi Oi 

(<^2^i{U9M{VJ ) ?unrr/JiKtut3X 



in a lymphatic region, although occasionally these are limited 
to the one physical condition of tenderness. Although elephan- 
toid fever has been classified as an entity separate from lym- 
phangitis, it is a question whether there is not in every case 
an affection of the deep or superficial lymphatics. 

During these attacks there is often severe headache, delirium, 
and anorexia. Predisposing causes to an attack are a severe 

19.3 Maxwell: Filariasis in China 271 

strain, as seen in the case of the burden bearers of this dis- 
trict; but overfeeding is a still more potent cause. During 
four years I frequently had, as one of my burden bearers, a 
man whose blood swarmed with filarise. Sometimes he used to 
visit an island where the people were most hospitable and fish 
was plentiful; as often as he visited this place, this man got 
an attack, apparently through overfeeding. Cold and wet do 
not seem to be very potent causes in bringing on an attack. 

As to treatment, there is no need to order the patient to bed, 
for he is feeling too ill to wish to be anywhere else. When an 
attack is well under way, there is no form of treatment that 
will stop it ; but if the case be seen before the fever has started, 
it may be partially or wholly aborted by a sharp purgative and 
a stiff dose of quinine. Three or four '^Livingstone rousers'' 
and 15 grains of quinine in acid are a good prescription, and 
the fever has been cut short by this method in a remarkable way. 

If there 'is a manifest accompanying lymphangitis, this should 
be suitably treated. Phenacetin sometimes gives relief to the 
headache and diminishes the fever, but its action in this respect 
is not at all certain. It may be said, generally, that once 
the fever is established, expectant treatment is the only course 
open to the medical man. 

There is but little practical difficulty in the diagnosis of 
elephantoid fever. As a rule, there is some accompanying 
lymphangitis; and even if there is none, generally the native 
is quite able to distinguish between it and malarial fever. On 
one occasion a man walked into the consulting room and told 
me that he was suffering from both filarial fever and malarial 
fever. He was quite right, his malarial attack taking place at 
the time he foretold it, the quartan parasite being easily found, 
and two attacks of filarial fever occurring during his fortnight 
in the hospital. This case had no manifest accompanying lym- 


In the great majority of forms of filarial disease, lymphangitis 
is present at one time or another. If the affected lymphatics 
are on the surface, the characteristic red streak on the skin 
and the tender cordlike swelling of the lymphatics are very 
manifest almost at the commencement of the attack. Often, 
even before the commencement of the fever, the lymphatic glands 
of the implicated region are a little swollen and tender. If the 
attack is at all severe, the inflammation will spread to the sur- 
rounding parts, the skin of which will become tense and shiny. 

272 '^he Philippine Journal of Science 1921 

Should the inflammation proceed farther, a severe cellulitis may 
ensue, or lymphorrhagia may take place. 

As a rule the inflammation subsides under appropriate treat- 
ment, leaving behind, however, a legacy in the shape of some 
permanent thickening. This thickening is Variable in amount, 
and it is almost impossible to predict in any given case what 
measure of thickening will be left. 

The area of inflammation is sometimes very extensive. In 
one case the lymphatics of both cords, superficial lymphatics of 
the scrotum, and the lymphatics of the upper parts of both 
thighs were affected, and the condition of the patient was by 
no means enviable. But in other cases the inflammation of the 
lymphatics is confined to a very narrow area. In the upper 
arm this is often the case, and the inflammation in one patient 
was confined to an area less than 2 inches (5 centimeters) in 
length in the region of the brachial artery. 

I have seen filarial fever without manifest lymphangitis; I 
have never seen filarial lymphangitis without fever of greater 
or lesser degree. 

Diagnosis is easy, as a rule, as the absence of local cause 
and the presence of filarise in the blood are readily determined. 
If there is a manifest possible local cause, it is sometimes dif- 
ficult at first sight to be sure, but before long the course of 
the case gives clear indication. 

As to treatment, rest, lead and opium lotion, and the treat- 
ment previously advised for elephantoid fever quickly clear up 
the trouble. 

If the attack fails to resolve, a careful watch, must be kept for 
the formation of abscess. It has been recommended that the 
swollen area should be scarified or pricked. Experience leads 
me to advise strongly against this procedure. It is extremely 
difficult to keep the parts about the scrotum, especially in 
natives, aseptic ; and it is quite possible to start a most trouble- 
some lymphorrhagia. In a few cases, after the inflammation 
has subsided, it may be advisable to dissect out the thickened 
lymphatics, and sometimes in this way it has been possible to 
secure the parent worm, which by its presence may have given 
rise to the lymphatic inflammation. 


It becomes necessary to use a name such as erysipelatoid to 
describe erysipelatoid inflammation, for the term ^'erysipelas" 
would convey a false idea. While it is no doubt true that the 
superficial resemblance between this affection and erysipelas 

19,3 Maxivell: Filariasis in China 273 

is by no means slight, yet there are very important differences 
to be noted. Foremost among these is the practically non- 
contagious nature of this disease. In earlier practice it was 
feared that Chinese students might carry infection from such 
cases to operation cases, but it was soon found that this danger 
was a negligible quantity, as cases may lie alongside newly 
operated cases without any harm occurring. In spite of this, 
however, it is not an advisable thing, as it is quite possible 
for a filarial subject to get ordinary erysipelas, and the distinc- 
tion is not always easy. 

The rash, too, is far more diffused and, as a rule, it has no 
clearly marked, raised border. Argument from the* constitu- 
tional symptoms cannot be used, as these vary within wide 
limits. It is not at all uncommon to see several of these af- 
fections of filarial origin present at the same time in one and 
the same patient, ^thus preventing any definite deductions from 
constitutional symptoms. Generally the diagnosis from ery- 
sipelas is easy because of these concomitant filarial affections; 
but occasionally a patient will come into the hospital with ery- 
sipelatoid inflammation and cellulitis of one leg, in whom this 
is practically the first serious symptom of the disease, and in 
such a case it is almost impossible to make an absolute diagnosis 
from erysipelas at first sight. 

I have not found it possible to follow the Brazilian (5) phy- 
sicians in dividing erysipelatoid inflammations into several 
classes. The only division that seems to me to be at all useful 
is to distinguish two forms: 

a. Erysipelatoid inflammation of all degrees of severity from simple 
erythema to an acute type to be described later, and which gen- 
erally terminates in the death of the part. 

h. Ordinary erysipelas in a filarial subject. 

The form in which the joints are simultaneously attacked is 
dealt with in a separate section. 

What is the materies morbi in these cases? 

In the case of h undoubtedly the poison is Streptococcics ery- 
sipelatis, and the pathology of these cases is perfectly well 

In the case of a, however, the pathology is not so clear. 
There is, undoubtedly, lymph stasis in the infected area, and 
bacteriological examination of these cases is much needed. Such 
work as has been done is not conclusive. There is evidence to 
show that it is not merely a case of poison introduced from the 
outside. If a lymphatic fistula becomes established in an af- 

181052 2 

274 "^he Philippine Journal of Science 1921 

fected region (and very often a leg ulcer may serve the pur- 
pose), no attacks of filarial fever or erysipelatoid inflammation 
will occur as a rule, so long as drainage is kept up. Heal the 
ulcer or close the fistula, and the attacks recommence. This 
lends some weight to the view that these filarial inflammatory 
attacks are due to a poison that is generated in the body itself 
and not introduced from the outside into tissues that are merely 
weakened by the presence of the parasite or into an area of 
lymph stasis. 

It is very easy to. say that all these attacks are set up by 
small wgunds, scratches, insect bites, and the like, but the 
evidence on this matter is far from satisfactory. Careful ex- 
amination has often been made for a point of local infection; 
but, although sometimes points that might serve can be found, 
it is difficult to satisfy oneself that such was the actual point 
of infection. 

Treatment consists of rest in bed and elevation of the af- 
fected part. In the case of the leg, wrapping it in a continuous 
cold-water dressing gives great relief in some cases. The 
bowels should be kept well open, and quinine and iron should 
be given internally. One of the marked features is the linger- 
ing character of some of these cases. An attack may last only 
a few days; on the other hand, it may drag on for a fortnight 
or three weeks. Not infrequently second and even third at- 
tacks may supervene while the patient is in bed recovering 
from the first. 


The section on dermatitis and cellulitis need only be dealt 
with shortly, as it concerns a more acute form of the inflam- 
mation seen in erysipelatoid inflammation or lymphangitis. 
Clinically it is just like an ordinary cellulitis, only much more 
amenable to treatment. For this reason it is not necessary to be 
in a hurry to incise, as many cases will yield to the treatment 
described in the last section. An additional reason for being 
cautious is that wounds made under these circumstances heal 
very slowly. A simple incision in a case of this kind may take 
seven months to heal. On the other hand, if it is clear that 
there is imminent danger of extensive sloughing, then free 
incision should be made at once, as the efl^ects of extensive 
sloughing are most disastrous. As an illustration may be ad- 
duced the case of a man who came under care in the last stage 
of this trouble. He was greatly emaciated; he had had cel- 
lulitis in the lower part of both thighs, with the result that 

19,3 Maxwell: Filariasis in China 275 

both legs were the subject of bad contracture. The cellulitis 
had also left him with lymphatic fistulse on the outer surface 
of each thigh. Both legs were the subject of elephantiasis, and 
the condition of the man was miserable beyond description. 
He was extremely anaemic and died of exhaustion ten days 
after entering the hospital. 

In rare cases the dermatitis becomes ulcerative. Of this 
only one good instance has been seen by me. It is always 
the result of infection of the bullae which are apt to form on 
the surface of a limb that is already affected by erysipelatoid 
inflammation or the like. It must be treated like any ordinary 
ulceration, its pathology being precisely the same. 


Abscess of the scrotum is a disease by no means common in 
England even when we include tuberculous abscess connected 
with the epididymis and testicle, and it was a cause for some 
surprise and not a little incredulity to be confronted, soon after 
arrival in China, by a patient whose scrotum, swollen to the 
size of a fcetal head, appeared to be little more than a bag 
of pus. 

The history of this patient was a curious but at the same 
time a typical one, although at the time it was not known 
to be such. Fifteen days previously, while at work, he had 
been seized with a violent rigor, which lasted about ten minutes 
and then passed into fever, which had been persistent since. 
He was cognizant of no previous illness, and denied any attacks 
of lymph fever, stating that he had been a strong man all his 
life. He was aged 42 years. He refused to come into the 
hospital, and I was unwilling to operate in his dirty hovel. 
Eventually part of the scrotum sloughed, and after a long illness, 
the man recovered. He was, however, subsequently troubled 
by attacks of elephantoid fever. 

In seeking for the cause of this and subsequent cases of a 
like nature, gonorrhoea could easily be excluded, as not more 
than 30 per cent of the patients confessed to having had it. 
None of them had stricture of the urethra, and the orchitis, 
when present, was slight; in many cases it was absent. Besides 
the fact that the abscesses were mostly outside the testicle and 
seminal tract, the further knowledge that Morris ^ did not men- 
tion such an affection put it almost out of court. 

^ Diseases of the Urinary and Generative System. Cassell, London 

276 ^^^ Philippine Journal of Science 1921 

Injury, with the suppuration of consequent hsematoma, 
next suggested itself as a possible cause. But in every case in 
the list injury could be absolutely excluded. 

Infection by the bite of some insect could also be excluded. 
Manson(27) mentions abscess as a manifestation of filarial 
disease, but it is well to realize that this may be the first and 
only manifestation. Systematic examination of the blood of 
the patients who suffered in this way showed every one to be 
suffering from Filaria bancrofti; and, in considering the sub- 
ject of filarial abscess, every doubtful case (that is to say, 
every case in which lymph fever could not be found or a definite 
history obtained of this symptom) has been excluded. Only 
about one in ten abscesses judged to be filarial in origin has been 
excluded owing to this test, and none of these were abscesses of 
the scrotum. 

To turn to the general question of filarial abscess, its inci- 
dence is naturally limited only by the incidence of infection 
with the filarial parasite, and it may occur either as an inci- 
dent in the course of filarial disease or as the first symptom 
and sign of the same. It may occur in any situation where 
there is loose connective tissue, rich in lymphatics, and for this 
reason the majority of abscesses outside the scrotum are in the 
vicinity of the great vessels, and the abscess does not always 
form in the situation where the inflammatory focus starts. 
In two of the cases in which a filarial abscess was opened over 
the lower end of Hunter's canal, there was good evidence that 
the inflammatory focus was first situated in Scarpa's triangle. 
Both the patients themselves and students who saw the cases 
before me are perfectly clear on this point; this evidence is 
important as bearing on the question of pathology. 

Previous observers, on opening such an abscess, have found the 
dead body of a parent filarial worm, and they have justly sur- 
mised that some at least of these abscesses were the result of 
the death of the parent worm. 

In one of my cases portions of the body of a parent worm 
were found, but in none of the others could anything of the 
kind be found, although a most thorough search was carried 
out. In some, however, the abscess was of some standing, 
rendering the absence of a dead parent worm insufficient proof. 
On two occasions broth cultures were inoculated without any 
result, and in, some there are undoubtedly ordinary pyogenic 
organisms, but these are cases which have been neglected and 
in which the skin is about to slough. 

19,3 Maxwell: Filariasis in China 277 

Dr. J. H. Montgomery has also found a parent worm in a 
case of this nature that came to him at Choanchiu, Fukien. 

Some cases are probably due to causes other than the death 
of a parent worm. I was fortunate enough to be able to ob- 
serve the whole process in a patient, who was in the hospital 
for a trivial affection. Within the space of a few days he had 
an attack of filarial lymphangitis of the cord on both sides of 
the scrotum; one side came to suppuration, the other did not. 
Both, at the outset, presented exactly the same appearance, 
the inflammatory process on the one side being apparently 
much the same as on the other. 

On two other occasions, by promptly putting the part at rest 
and applying cold, the inflammatory process has been aborted. 
As a rule, a sharp local attack leads to the formation of pus. 

There is another fact that has to be taken into account in 
considering the pathology ; namely, that some of these abscesses 
in the scrotum are suppurating hydroceles. In such a case it 
is difficult to see where the death of a parent worm can come 
in, as parent worms have never been found inside hydroceles. 
One is inclined to think that the majority of these are due to a 
local cause and are probably connected with the blocking of lym- 
phatic vessels ; but there is no evidence to offer in support of this 
view save the fact that in some cases the abscess is the first 
manifestation of an affection that ends in elephantiasis of the 
scrotum or of the affected limb. In others it certainly does not 
end thus, but possibly this is due to the fact that a lesser area 
of the lymphatic system has been interfered with. 


Passing from the pathology of the disease to its clinical form 
and diagnostic point, these abscesses may be classified as 
follows : 

Filarial abscess: 

A. Of the scrotum. 

1. Suppurating hydrocele. 

2. Abscess of the cord. 

3. Abscess below the testicle. 

B. Of the limbs. 

C. Intra-abdominal or intrathoracic. 

Concerning the cases contained in A, the division may seem 
at first sight to be arbitrary and unnecessary. A further con- 
sideration will show that it is not really so. Commencing in 
the way that is common to all abscesses, they follow different 
courses and each must be treated in a different way. 

278 ^^^ Philippine Journal of Science 1921 

Suppurating hydrocele. — A hydrocele of moderate or large 
size is usually already present. At the commencement of the 
attack this swells to nearly double its previous size and becomes 
intensely tender and painful. In a day or so the contents are 
purulent, and if left to itself the wall and skin over it will 
slough and the contents will discharge. A long convalescence 
may ensue, as the opening thus formed may be very large. 
Tapping is of use only in the early stage, and when the contents 
have become purulent, incision and drainage should be employed. 
Gare should be taken to have the opening at the most dependent 
part, and the thick layer of lymph lining the hydrocele sac should 
be removed. Free bleeding will ensue, and the cavity may have 
to be packed for twenty-four hours with gauze. Healing is 
usually rapid, but care should be exercised not to allow the ex- 
ternal opening to close too soon. Strict antiseptic precautions 
should be observed. 

Abscess of the cord. — This is the most serious of the forms 
of abscess affecting the scrotum, by reason of the liability of 
the inflammatory process to spread up the cord. In several 
cases I could trace the cord as a thick, hard rope as far up as 
the internal ring; moreover, it was intensely tender. Whether 
or not in these cases there is any associated inflammation of 
the neighboring peritoneum, there is no positive proof; but the 
fixity and tenderness of that part of the abdominal wall is sug- 
gestive, and it is well known that an acute septic infection of 
the cord may spread inward and set up an acute septic peri- 
tonitis. Although the inflammation in these filarial cases may 
spread up the cord, the actual abscess in all the patients seen 
was outside the external abdominal ring. Abscess of the cord 
is also the most serious of the forms of abscess affecting the 
scrotum as regards treatment. The abscess is apt to have loculi 
and pockets, which interfere with free drainage and, consequent- 
ly, with rapid healing. 

Abscess below the testicle. — This is the simplest of the filarial 
abscesses occurring in the scrotum. Drainage is easy, and heal- 
ing is rapid, a few days sufficing to end the affection, as the 
abscess does not extend far and is not of large size. From one 
of these the portions of a parent worm were obtained, and it 
is possible that all the abscesses in this situation are due to a 
similar cause. 

Filarial abscess of the limbs. — ^As I have previously stated, 
abscesses of this kind occur in situations rich in lymphatic 
tissue, and generally in the immediate neighborhood of the 

19,3 Maxwell: Filariasis in China 279 

large vessels. It is not an uncommon thing, on opening one of 
these abscesses, to be able to put a finger on or even around the 
main vessel of the limb. It is one proof of the mild form of 
inflammatory process that I have never seen one of these vessels 
give the slightest trouble from secondary haemorrhage, due to 
softening of the vessel wall and subsequent rupture. 

Of these abscesses, those in relation to the femoral artery give 
the most trouble. This is partly on account of their deep con- 
nections and the difficulty of thorough drainage. In some of 
them the lymphatic system seems to be so thoroughly dis- 
organized that they pass directly into the commencement of an 
attack of elephantiasis of the limb. Consequently prognosis of 
the ultimate result of treatment must be guarded. 

The treatment consists in free drainage; owing to the dif- 
ficulty of draining the deeper portions, it is well to make free 
openings under chloroform and insert large rubber tubes. 
Troublesome contracture of the lower limb may occur during 
the healing of large abscesses, especially those involving the 
popliteal space. This can be avoided by splinting the limb and 
using massage as soon as possible. But among Chinese pa- 
tients, who are absolutely intolerant of restraint, and over 
whom we have not the same command as in England, trouble- 
some contracture is by no means unknown. 

Intra-abdominal or intrathoracic abscess. — I have had no 
experience with intrathoracic abscess, but have met with four 
cases of the intra-abdominal form. In each case the illness coin- 
cided with the cessation of attacks of elephantoid fever, and 
each case was desperately ill when admitted to the hospital. 
All had commenced in the way hereafter described as typical. 

In two cases deep fluctuation was present in the left lumbar 
region, and under chloroform a postperitoneal abscess was 
opened that apparently had no connection with any of the large 
abdominal organs. The urine contained the faintest traces of 
albumen in one of these cases, but neither formerly nor at the 
time of operation were there any symptoms pointing to disease 
of the kidney. 

The first patient left the hospital. at his own request; he was 
still desperately ill. The abscess was draining well, however, 
and he made an excellent recovery after having been ill for two 
and a half months. 

The second was a similar case. He was operated upon in 
the same place, but made a speedier recovery. 

280 ^^^ Philippine Journal of Science 1921 

The third came into the hospital looking like a typical ap- 
pendicitis patient. He also was very ill, having been so for 
one and a half months. The abscess proved to be extraperito- 
neal; it extended both into the iliac fossa and downward into 
the pelvis. Probably it began in the loose tissue about the iliac 
vessels; the fact that the disease began with a rigor and im- 
mediate flexion of the right thigh on the abdomen to an extent 
that is rarely seen in an uncomplicated attack of appendicitis 
lends color to this supposition. This case slowly healed up, 
and the man regained health and strength. His bowels were 
regular and natural at the time of the attack. 

The fourth was a case of perinephric abscess that had been 
neglected and had tracked into the gluteal region. Free inci- 
sions had to be made in both the lumbar and the gluteal regions. 
This patient was also very ill, but with free drainage and care 
he, too,* did well. 

Hanson's (28) advice on the subject of these abscesses is sound : 

Deep seated pain in the thorax or abdomen, with inflammatory fever 
followed by hectic, and a diminution in the number of micro-filiarise in, 
or their entire disappearance from the peripheral blood, should, in such 
circumstances, suggest a diagnosis of filarial abscess, and indicate explo- 
ration, and if feasible, active surgical interference. 

Finally we must discuss the onset of this malady and its 
typical temperature. Fig. 2 is a typical temperature chart of 
the disease. Rising rapidly with a rigor, which may occur 
while the man is at work in the fields, cutting wood, or even 
lying in bed, it remains high for from one to three days and 
then descends by lysis, provided an unopened abscess is not 
left. On the other hand, if the patient be under favorable condi- 
tions and it is possible to abort the attack, the temperature may 
descend very rapidly (fig. 3). 

In neglected cases and cases that do not run a proper course, 
owing to insufficient drainage or wide infiammatory focus, the 
temperature may assume a septic type (fig. 4). 

In criticism it may be urged that this temperature is the 
result of Plasmodium malaria complicating the filarial attack. 
In reply it may be stated first, that in many of the cases the 
blood was carefully examined for Plasmodium malarix with 
negative results; and, second, that it is quite possible for the 
fever, due to the presence of Filaria bancrofti, and active mala- 
rial attacks to run concurrently, but in such cases both diseases 
are clinically and microscopically distinct; and moreover the 
patient, as I have proved in two instances, has been right in 


Maxwell: Filariasis in China 


his statement that one night he had a filarial attack and the 
next an ordinary quartan paroxysm. In no case of filarial 
abscess of the limbs referred to in these lists was there any 
focus of infection to be discovered on the surface of the limb. 


A great many diseases in times past were credited to malaria. 
Some of these seem to have been so considered on totally 





















NorrruxJj \ 


otBodff J 













n \ 














1 1 


\ ■ 



-N — 1 



\ / 







- *> 
























\ J 






'I '.♦'a? 



■— ■ 























Fig. 2. Typical chart; filarial abscess. 

inadequate grounds. In fact, the presence of malaria in a 
country has been held to account for almost any obscure disease 
in that country. While not denying that there may be a form 
of malarial orchitis, it is quite clear that in the region in which 
I have been working the majority of cases of orchitis are not 
malarial but filarial. In fact, I have never ceen a case in which 


The Philippine Journal of Science 


there was any suspicion of its being malarial ; and this in spite 

of the fact that the whole region is scourged with malaria, large 

spleens being common, and malignant cases by no means rare. 

Here is a typical example of a case of filarial orchitis: A 















1 ^^' 


1 Date 


•••-- - 








^"" ""; 








- — 



I f 


-^^ •■ - - - 

— • 1 




0- ■ ■ - 




1 .J 



F cur n 




t- - 



I J 




r 6 


2>< yrfy( 







,. _ 

L ""±3""' 




- f'"T" 


f ^ 

1 ^.i 






.._. . 


:^ ■'"' 




■■-•i ■-■••■■■•'- 


:s ■ 

:: z 




re s 



•" s 7 •■;■• 





— -€ ~" ■ 

















)d w 

2; 2% Z3 






Fig. 3. Filarial abscess. 

man of 45, who had previously suffered from filarial abscess 
and had lived a clean life, having had neither gonorrhoea nor 
syphilis, was suddenly seized with an attack of elephantoid 
fever. Filarise were swarming in his blood at the time. On 
the next day his left testicle was swollen and intensely painful, 


Maxwell: Filariasis in China 


the skin of the scrotum was red and swollen on the same side, 
and the constitutional symptoms were severe. So wretched did 
the man feel that he assured me he was going to die. The acute 
stage soon passed away under treatment, but the trouble was 
not quite cured for a month. In this case there was also lym- 
phangitis of the spermatic cord. He has had one attack since, 




i ? t J 




1 j : : 1 . : : ! 








■• ! 

! 1 \V 

, , 1 ,4 . 


! l|j 


in i 

"V\\ \ 






1 : ; : 

1 i < I 







t r 1 1 






\\ vh 


i \ \ ' 














: ! : J 

1 ' .; J 

' 'l+H 



I j 



' ' * 1- 


1 1 i 

! 1 i I 


1 1 




i J f i 

, I. 


' ' i I 


"• • It 

i ; i 1 







1 i ' 1 

i • 1 1 



j : I i 



' ' ! 1 

; ; i ; 




' 1 ; ; 





! 1 I 1 

: i j 

' \Ji^ 


i '! ri 

I 1 i I 

1 1 < 

' ! • ! , 


i j 

,' ' i.L 





1 i ! : 

; 11 ; 

' 1 ' f 

1 ■' ; 1 
; ; 1 1 

»4 ; 



i ' i ! 

i : ; ! 


; : 





! 1 ! 1 

1 I " I 


; i 


f— ' 


I ! 1 -r^ 



1 1 1 1 



1 1 j i 



' ' ' 1 

i . 


: 1 : 
1 ' • 










■4- 1 !-f- 

: ; : ' 

[ ! 



; : • : 


VJ , 1 1 

i ' 'i ' 

; r: i 

1 i 


1 : I ! 




; i ; : 


1 1 ; ; 

! 1 1 I 

] I 

; ; ! 1 


; 1 1 i 


■; i ! 1 


; : ; j 






1 : ! j 

! ; 



\ • :^ 


1 ' ' ' 

; ; ' i 

! I ' ! 

1 ; 


: ; ! ! 





; : ! : 


: i I i 

! 1 ; : 

1 ; 





! 1 : ! 

1 1 i 1 









i i"! : 




' ' i ; 




1 • I '. 1 




' '^M t^ 



• 1 . ! 






i^c^di \s 


\ ' 

! i ' 1 



'^ CO cv^ "^ © a> °go gr^ 

2 2 2 2 2 « =^ ^^11 

and the symptoms were about the same. Six months separated 
the two attacks. 

In four years' practice in the coast belt, three well-marked 
cases of this kind have been seen. Nearly always when there 
is filarial abscess of the scrotum there is some accompanying 

284 The Philippine Journal of Science 1921 

orchitis, but the orchitis is secondary; it is not very severe, is 
anaenable to treatment, and shows no tendency to recurrence. 
Tendency to recur is marked in the true uncomplicated form. 
In some cases it leaves behind it a thickening of the spermatic 

Its pathology is probably the same as- that of the other 
diseases in this section ; that is, inflammation in an area of lym- 
phatic congestion. In operating on a case of filarial lymphatic 
varix, Moty(38) discovered several soft cysts on the surface of 
the testicle which was enlarged. There was also great thicken- 
ing of the spermatic cord in this case. 

The treatment of the disease is the same as that of ordinary 
orchitis, and strapping should be employed if the case becomes 
chronic. If there is much pain, puncture of the testicle with 
a fine tenotome gives great relief. 


In cases of erysipelatoid inflammation, affecting a limb, for 
example, it is by no means uncommon to hear a complaint by 
the patient of pain in one of the proximate joints and, on 
examining the same, to find evidence of some synovitis present. 
As a rule, this synovitis is a negligible quantity and gets better 
without treatment; but in a few rare cases the inflammatory 
process is much more severe. The connection between acute 
synovitis of the knee joint and filarial infection has been pointed 
out by Maitland.(2l) My own cases consisted of two kinds. 

a. Cases where the infection almost certainly came from the outside 
and simply fastened itself on a previously inflamed joint. 

h. Cases where the infection was part and parcel of the inflammation 
depending on the presence of the parasite. 

As an illustration of the first kind, the following is of in- 
terest : A man with an old lymph scrotum, who had previously 
had erysipelatoid inflammation of the leg, with accompanying 
pain in the knee, came into the hospital for some troublesome 
internal haemorrhoids, of which I ligatured two masses. He 
was operated upon in the morning. The same evening he had 
a smart attack of erysipelatoid inflammation of the right leg. 
Next morning he was complaining of his knee joint, which was 
distended with fluid. On the following day he was worse, and 
the joint was aspirated and turbid fluid was drawn off. This 
did not clear up the trouble, and two days later the joint was 

19.3 Maxwell: Filariasis in China 285 

opened and drained. He made a good recovery with a movable 

One more case of this kind has come under my care. The 
same joint was affected, and the history was identical. In this 
case the affection followed an operation for fistula in ano. Re- 
covery followed drainage of the joint, but the movements of 
the joint were far from perfect. 

These cases are the only two, in a hospital practice of twenty 
years' duration, where infection of a joint has followed an every- 
day operation, and it is curious that both should have occurred 
in the subjects of filarial infection, filar ise being easily found 
in the blood. The usual antiseptic precautions were taken in 
both cases. 

In regard to the second kind — ^that is, cases where the infec- 
tion was part and parcel of the inflammation depending on the 
presence of the parasite — strange to say, in both cases seen 
the joint affected was the wrist joint. The first case was that 
of a man 40 years of age. He had a lymph scrotum, and his 
blood swarmed with filarise. On February 21, 1902, he was 
attacked with erysipelatoid inflammation of the right arm and 
forearm. Three days later his right wrist joint had become 
very painful. He was treated in the usual way, and the inflam- 
mation subsided except around the wrist joint, which remained 
swollen and very tender. On March 6 he was put under chlo- 
roform, and the joint was explored through a palmar incision. 
A little turbid fluid came out, and the ends of the bones enter- 
ing into the joint were rough. The incision was then closed, 
and the part was put up in plaster of Paris. This was kept 
on for a month, and then massage was applied. The patient 
made a perfect recovery, the movements of the joint being normal 
when he was last seen. 

Guided by the last case the treatment of the second patient 
was modified considerably. He was a man of 50, who had 
suffered from lymph scrotum for twenty-five years. He was 
attacked in the same way as the first patient — if anything, more 
severely. There was marked grating in the joint, which was 
intensely painful. After the acute symptoms had subsided, the 
part was put up in a fixed apparatus, and the joint was soon 

Acute rheumatism is not present in this region, and both 
patients were otherwise free from disease. 


The Philippine Journal of Science 


Gangrene of the scrotum is not a common disease, especially 
if all cases of nephritis are excluded. Its occurrence in a healthy- 
young man is sufficiently remarkable to call for special notice. 
Table 7 gives particulars of 8 cases seen during four years. 

Table 7. — Cases of gangrene of the scrotum. 


O. T-__ 
O. H^a- 
T. L .._ 

C. K- 
T ... 
B ... 



Evidence of filarial disease. 


Filarial embryos and varicose groin glands Yes. 

Old elephantoid fever; no embryos found Yes. 

Varicose groin glands; fever; no embryos found Yes. 

Elephantoid fever and occasional swelling of scrotum; no em- No. 
bryos found. 

Varicose groin glands; elephantoid fever __. j Yes. 

Elephantoid fever; no embryos found No. 

Varicose groin glands; no embryos found Yes. 

Elephantoid fever for many years; no embryos found ; No. 

" Slight erysipelatoid inflammation of the scrotum accompanying attacks. 

The only common feature in these cases is that all had a history 
of filarial disease, past or present. They began in the same 
way; that is, with a sharp attack of fever, accompanied by 
redness and swelling of the lower half of the scrotum. This 
inflammation is from the first acute, and in a few days the lower 
half of the scrotum becomes black and comparatively dry. If 
left to itself, this gradually separates, and the exposed surface 
slowly heals. Healing may be hastened by skin grafting as 
soon as the surface is covered by healthy granulations. 

Hydrocele, if present, may materially hinder the healing of the 
wound, and it must be tapped. If this is done with strict anti- 
septic precautions, there need be no fear of infecting the hydro- 
cele. It should be tapped, of course, through healthy skin and not 
through the granulating surface. 

It may be urged that these were malarial cases. While it is 
true that malarial gangrene is a well-known affection, in all 
these cases no evidence was obtained that would support this 
contention. Although none of them had been treated with qui- 
nine, the microscopical examination of the blood was entirely 

One point is worthy of note; namely, that of the eight cases 
only two had filarise present in the blood. It is possible that in 
all such cases, however, the parent worm is present in the tissues 

19.3 Maxwell: Filariasis in China 287 

that die, and that the death of the tissues is due to the inflamma- 
tion set up by the death of the worm. 

Another point worth noting is the little constitutional dis- 
turbance manifest in the patient after the initial sharp attack is 

It may be urged that the evidence in favor of its being due to 
filarial infection is too slight. There is an important negative 
piece of evidence. I moved from Changpu, in the coast belt, 
to Yungchun, which is farther inland, and pari passu with the 
loss of practically all filariasis cases; gangrene of the scrotum 
became such a rarity that only one case was seen in twelve years, 
and the patient had resided for some time in the coast belt. 


By filarial haemoptysis is meant haemorrhage from the lungs 
occurring in patients with filarial infection, correlated in time 
with an attack of filarial fever, with or without manifest evi- 
dence of lymphangitis. 

In the case specially studied by me, there was a heavy filarial 
infection, and there was no manifest evidence of lymphangitis. 
It is known that during the day the filarial embryos have their 
habitat in the lungs ; in the case that first directed my attention 
to the subject the haemorrhages took place during the day, and 
live filarial embryos were present in the expectorated blood. 

It may be said that the Chinese are very commonly subject 
to tuberculosis, and in the absence of post-mortems it is impos- 
sible to state that these haemorrhages are not tubercular in ori- 
gin. The only answer I can make is that the case here discussed 
has been under observation from time to time for the last twenty 
years; that the haemorrhages only occurred when he was the 
subject of filarial infection (he became free from infection for 
several years during this time) ; that the sputum was carefully 
examined for tubercle bacilli many times with negative results ; 
that examination of the chest was consistently negative ; and that 
the patient never presented any symptom, save the haemorrhage, 
that would suggest a tubercular infection. 

I have never seen one of these cases die of the haemorrhage. 
It may be of fair quantity, but it stops with the fall of the tem- 
perature and does not recur apart from a definite attack of fila- 
rial fever. 

It is impossible to be dogmatic on the subject, but it is my 
conviction that the above diagnosis is correct, and this opinion 

288 The Philippine Journal of Science 1021 

is shared by both Dr. H. T. Whitney and Dr. J. H. Montgo- 
mery, of Fukien, who have had a large experience with filarial 


In commencing this section on disease due to obstructive inter- 
ference with the lymphatic system, it must be clearly understood 
that this division is an arbitrary one. While it is true in the 
main that the diseases of the one set are chiefly the result of in- 
flammation, it must be borne in mind that there is at the same 
time, as a rule, lymphatic stasis in the affected part. On the 
other hand, although we speak of disease due to obstructive inter- 
ference with the lymphatic system, it must be remembered that 
in the production of the lymphatic obstruction inflammation 
plays an important part. This point being clearly understood, 
we can turn to the consideration of details. 

There are two main forms of filarial disease to be discussed 
under this section. Roughly speaking, the one is characterized 
by dilatation of the lymphatic, the other by the production of 
more or less solid oedema in addition to this dilatation. One 
of the main proofs that all of these diseases are due to the same 
cause is found in the fact that all grades of disease may be seen, 
from the small lymphatic varix to the most solid forms of 

How do these diseases originate? It cannot be said that it 
is always clear, and each case will be discussed in its proper 
section. The parent worm or worms may act as an embolus in 
one of the larger lymphatics, and may cause an attack of in- 
flammation, or bleeding into the lumen, ending in the stenosis 
or occlusion of the vessel. The ova of the worm, if discharged 
prematurely, are probably able to block the smaller lymphatics, 
and are also able to block the circulation through the lymph 
glands. In either case a portion of the lymphatic system be- 
comes wholly or in part cut off from the general lymphatic cir- 
culation. The pressure in this area rises, and varicosity of the 
lymphatics, or lymphatic oedema, or one of the many combinations 
of these two, results. 


The ramifications and anastomoses of the lymphatic system 
are fortunately very free, so that a compensatory circulation is 
not very difiicult to establish. But in any case of lymphatic 
obstruction (especially if large trunks, such as the thoracic 
duct, are occluded) it takes a little time for this compensation 

19.3 Maxwell: Filariasis in China 289 

to take place. Meanwhile the lymphatics of the occluded area 
are dilated, and lymphatic varix occurs. If the thoracic duct 
is occluded, then the chyle can only reach the general circulation 
by a retrograde route, and in consequence the abdominal and 
pelvic lymphatics become much dilated. Manson(29) speaks of 
these in the following terms : 

In dissections of such cases the thoracic duct has been found distended 
tc the size of a finger, the abdominal and pelvic lymphatics forming an 
enormous varix, perhaps a foot in diameter and many inches in thickness, 
concealing kidneys, bladder, and spermatic cords. In such cases, when one 
of the vessels of the varix is pricked or ruptures, the contents are found 
to be white or pinkish. They are not limpid like ordinary lymph, they 
are chyle, therefore, chyle on its way to enter the circulation by a retro- 
grade compensatory track. 

Lymph scrotum, varicose groin glands, chyluria, and the like 
have as their basis a condition of lymphatic varix, but they are 
not pure lymphatic varices. They will be discussed under sepa- 
rate headings. 

There are, however, cases of pure lymphatic varix that we 
must discuss as such. Occasionally one meets with cutaneous 
lymphatic varix. There are small swellings which, although 
sometimes permanent, are often evanescent. The kind de- 
scribed as ''deeply-situated little swellings" (30) has been seen 
but once. On the other hand, I have sometimes met with 
groups of vesicles on the thigh, which might well fall 
under this category. They may occasionally form the starting 
point of a lymph fistula. The contents of these vesicles accord- 
ing to my experience are always clear fluid. 

The deeper lymphatic varices are rarely seen, owing to the 
utter impossibility of obtaining post-mortems in inland China, 
but on one occasion a lymphatic varix of the spermatic cord 
was encountered when operating for the radical cure of hydro- 
cele. In this case all of the lymphatics of the cord were 
dilated, and there were two or three small cystic dilatations on 
the lymphatic vessels. The contents were clear. No operation 
was performed on these dilated lymphatics, and the case con- 
valesced like an ordinary radical cure. The patient had had 
elephantoid fever, and his blood contained many filarial embryos. 

On another occasion I helped Dr. J. H. Montgomery to operate 
on a school boy, aged 16. In the right inguinal region there was 
a soft, compressible swelling, which enlarged when he stood 
up, and in which there was a marked impulse on coughing; 
on lying down it practically disappeared. The swelling turned 

290 ^^^ Philippine Journal of Science 1921 

out to be a lymphatic varix of considerable size situated in the in- 
guinal canal. The lymph vessels were the size of a large quill 
and ran lymph on being wounded. The whole mass was re- 
moved, the vessels being tied as if they were a mass of varicose 
veins, and the patient made a good recovery. 

On two occasions, when operating for elephantiasis of the scro- 
tum, marked dilatation of the lymph vessels about the spermatic 
cord was seen over and above the general lymphatic dilatation 
seen in these cases. 

As a rule, lymphatic varix is best let alone. If giving trou- 
ble, and if it is possible to remove the whole varix, this may be 
done; but it must be remembered that the varix is partly com- 
plementary, and any interference with it may involve the 
production of another varix elsewhere, or the formation of 
lymphatic fistula in some other part of the varicose area. 


Under the heading of lymphatic fistula there are three varie- 
ties to be described — the spontaneous, the inflammatory, and 
the operative. The first forms spontaneously in an area already 
in a condition of lymphatic varix. 

Spontaneous. — As an example may be adduced the case of a 
man, aged 23, the subject of filariasis. On the outer surface 
of the left thigh, rather toward the front of the limb and about 
its middle, was a small aperture from which lymph slowly 
drained away. A fine probe was passed in about a quarter of 
an inch (0.6 centimeter) inward and upward. There was no 
inflammation about the mouth of the fistula, the history of which 
was that it had formed spontaneously three days previously. 
Its appearance was preceded by a vesicle, and the lymph coming 
from it was transparent and free from blood. 

This form is rarely seen, and the mouth of the fistula remains 
free from redness or swelling, unless secondary infection takes 
place. Unfortunately with native patients, who love to iiieddle 
with any aperture or wound, this almost always occurs, and 
then it becomes a septic sinus. Unless it is giving much trouble, 
it is best let alone, as its presence probably saves the patient 
from many attacks of elephantoid fever. In one case I cau- 
terized and got the fistula to heal, but immediately the attacks 
of elephantoid fever, which had intermitted while the fistula 
was open, recurred. 

Another patient presented a small, uninflamed aperture on the 
front of a small lymph scrotum. At times lymph would spout 

19,3 Maxwell: Filariasis in China 291 

out of the aperture to a distance of 6 inches (15 centimeters) 
and a test tube full of lymph could be collected in a short time. 
The strain on the patient was considerable, and I stopped the 
flow by surrounding the aperture with a purse-string suture 
of the finest horsehair. During the patient's stay in the hos- 
pital the flow did not recur. 

The second variety follows the opening or bursting of a 
filarial abscess. It may for convenience be called inflammatory. 

Inflammatory, — It is not at all uncommon, after the opening 
or bursting of a fllarial abscess, for the patient to be troubled 
for some months by a lymph fistula in the site of the opening 
into the lymphatic abscess cavity. This sinus may be several 
centimeters deep, and it often runs down to the region of the 
lymphatics in connection with the large vessels of the limb. 
The usual site for a fistula of this kind is in the popliteal re- 
gion. The abscess cavity closes up, leaving a fine sinus, through 
which lymph gradually trickles. Another favorite situation 
for this kind of fistula is in an elephantiasis scroti, an abscess 
forming in the elephantoid tissue, bursting, and leaving a lym- 
phatic fistula. 

As to treatment, in both regions it is better to do nothing 
in the first case because, given time, the fistula will almost 
certainly heal of itself; and in the second because the probabil- 
ity is that nothing short of removing the elephantiasis scroti 
will be of any avail. In some cases this also heals spontaneously, 
but this rarely occurs. 

Under this class must also be included the ulcers of the leg 
occurring in elephantoid disease of this limb. These also act 
as fistulsB and are constantly moist with exuding lymph. They 
are very difficult to heal. A case is recalled of a man with 
both legs elephantoid and ulcers on one leg just above the ankle 
(Plate 21). As out-patient and in-patient I worked on him for 
three years, and finally gave up all treatment as hopeless. Al- 
though the ulcers diminished in size, they never showed any 
inclination to heal, and there was too much fibrous change, about 
the ulcer to make it possible to do. skin grafting with any 
reasonable hope of success. Besides, patients suffering from 
this affliction, although they wished to get their ulcers healed, 
yet confessed that the advent of the ulcer had almost entirely 
freed them from attacks of elephantoid fever. 

The third class comprises those following operations. 

Operative. — Temporary postoperative fistulas are by no means 
uncommon. Twice in operating on varicose groin glands has 

292 The Philippine Journal of Science 1921 

the operation wound healed by first intention, except for a fine 
aperture at the lower end of the scar, which continued to run 
lymph for from a fortnight to three weeks. In the end, how- 
ever, these fistulse heal satisfactorily, and they may safely be 
let alone. 

But it is sometimes otherwise with the fistulse that may be 
left after operations for elephantiasis scroti. Occasionally the 
elephantoid change extends high up on the skin of the abdomen. 
In this case, particularly if the patient is well on in years, it 
is better to leave some of this diseased skin rather than make 
a huge operation wound. But this carries with it its own pen- 
alty, as it becomes necessary to operate through diseased tis- 
sues, which are notoriously difficult to heal. Then a fistulous 
opening is apt to occur in the region of the root of the penis. 
Twice this has occurred in my own patients: in the one case 
the fistula healed spontaneously, in the other case it was still 
open a year after the operation. 


Where, in the course of the lymphatic varix, lymphatic glands 
occur, these participate in the general dilatation. In marked 
cases, on removal from the body, they miay appear on section 
like a sponge. In other cases, especially where there have been 
many attacks of elephantoid fever with inflammation of the 
glands, they may be indurated, and there may be a great in- 
crease in the fibrous tissue forming the framework of the gland. 
Of these varicose glands by far the most important are those 
found in the groins of patients suffering from filarial disease. 

These groin swellings may be found alone, or in association 
with other forms of filarial disease, notably with lymph scrotum, 
elephantiasis of leg or scrotum, chyluria, or chylous dropsy of 
the tunica vaginalis. 

In a not inconsiderable number of cases the patient is quite 
unaware of their presence until they are discovered by the 
medical man. They are often quite painless and give no 
trouble whatever. In other cases the patient's attention is called 
to them by dragging or aching pains in the groin. The swellings 
vary much in size, from a slight enlargement to masses the size 
of a man's fist. In many cases the glands are discrete, but in the 
majority the lymphatics running into the gland are also dilated, 
and the whole forms a mass. All of the groin glands may form 
a large, ill-defined, matted swelling, inguinal and femoral glands 
being alike involved. One side only may be the seat of these 
swellings, but very frequently both sides are affected. In many 

19.3 Maxwell: Filariasis in China 293 

cases the swelling is much smaller on one side than on the 

The skin over these swellings is natural. In the case of 
the smaller swellings it is nonadherent, but if there has been in- 
flammation about the glands, it may be difficult on operation 
to free it from the mass. The swellings themselves may be said 
to be adherent to the underlying fascia. A hypodermic syringe 
draws off lymph from these swellings, either clear or chylous, 
according to whether or not the varix is in connection with the 
chylous lymphatics. This fluid very often contains filarial 

As to their diagnosis, it has been said that it is a common 
mistake to confuse these with hernise. I have seen some two or 
three hundred cases, but have never seen one in which the 
slightest doubt on this head arose in my mind ; and,, with proper 
care in the examination of the swellings, this mistake should not 
occur. As there is often some other manifestation of filarial 
disease present, this may assist in diagnosis, if there is doubt. 
Masses of tubercular or lymphadenomatous glands might, if 
sufficient care be not taken, be mistaken for them, but there 
is one valuable positive sign that should greatly help: when 
the patient lies down with the buttocks raised, the tumors grad- 
ually get smaller, a condition that may be hastened by pressure 
with the hand; if the patient now stands up, the tumors 
gradually fill and become prominent. There is no impulse to 
speak of over these swellings. Manson(3l) adds a hint that is 
a very wise one : 

Chronic swellings about the groin, cords, testes, and scrotum, in patients 
from the tropics should always be regarded as being possibly filarial. 

Azema(i) has made the statement that these various glands 
may spontaneously disappear at or about the age of 40. It is 
difficult to see why this should be the case. I have never seen 
one spontaneously disappear, nor have I been able to get satis- 
factory evidence of this taking place. Certainly they generally 
do not, and many cases are seen after this period. Once I saw 
them appear for the first time in a woman of 64. She had 
come for dragging pains in the groins, accompanied by slight 
hsematuria, for which no cause could be discovered. The haema- 
turia ceased on the day of admission, leaving a urine in which 
nothing wrong could be discovered. Four months later she 
returned with markedly varicose groin glands, her blood full 
of the embryos, and it was clear that here was the explanation 

294 '^^^ Philippine Journal of Science 1921 

of her attack. Since that time she has had a distinct attack of 
chyluria, but not a severe one. 

As to treatment, I agree with those who have had experience 
of the subject that it is advisable to let these glands alone unless 
they are causing a large amount of trouble ; in troublesome cases, 
the glands may be safely removed. 

Septic lymphangitis used to be greatly feared in these opera- 
tions, and of course, unless care be taken, the operation is a 
dangerous one. No one who has not performed these operations 
can have any conception of the mass of dilated lymphatics, and 
these run when wounded and obscure the field of operation. As 
all the superficial veins and arteries are embedded in the mass, 
care must be taken, especially at the saphenous opening. The 
external saphenous vein is often a large one, and if cut just 
at the opening, it may be difficult to secure. As far as possible, 
the dilated lymphatics entering the mass should be ligatured, as 
this greatly diminishes the lymphorrhagia. The best material 
to employ for this purpose is fine catgut. 

It must always be remembered that removal of these glands 
may precipitate an attack of chyluria or elephantiasis of the 
leg. An appearance of the beginning of elephantiasis of the 
leg, however, must not be taken as a certain sign that this trouble 
is going to supervene. In one patient I was much distressed to 
find that, a fortnight after the operation, the leg presented every 
appearance of commencing elephantiasis. It was swollen, and 
with a solid oedema affecting the skin in parts. Six months later 
the leg was perfectly normal, and it was clear that the condition 
was only temporary, due to the collateral circulation not having 
become fully established. 

That the operation is very satisfactory in some cases is proved 
by the fact that the patient mentioned above came back in order 
to have a second operation on the opposite groin. 

Rarely these varicose glands are met with in other exposed 
situations. One case of varicose axillary glands has been seen. 
In this case the mass was not very large. 

In some cases the mass of glands forms the dwelling place 
of the parent worm. 

If the glands are not removed, the daily painting of the skin 
over them with linimentum iodi seems to exert a beneficial 
effect. How it does this is not clear, but it certainly in many 
instances relieves the aching and dragging pain of which the 
patients frequently complain. 

19,3 Maxwell: Filariasis in China 295 


In a well-marked case of lymph scrotum the appearances are 
characteristic. The scrotum is swollen, and the natural rugosity 
of the skin is increased. This appearance is partly due to the 
fact that, during the attacks of elephantoid fever, the parts 
become very oedematous and partly to the condition of the skin 
itself. The latter is a little thickened, and the surface is covered 
with small vesicles. The whole is part of an extensive lymphatic 
varix involving the scrotum ; and, as might be expected in these 
cases, the groin glands on both sides are often varicose, but 
this is not always so. In some cases it is only the scrotal skin 
that is affected. 

The character of the fluid obtained from the vesicles varies 
with the site of the obstruction. If the obstruction is affecting 
the chyle vessels, the fluid will be chyle or sanguineous chyle; 
if not, it will be straw-colored only. When an acute attack su- 
pervenes, the vesicles on the skin, otherwise quite small, enlarge 
and become tense. If very carefully handled, the attack may 
pass off without any discharge of lymph; but, as a rule, the 
vesicles either spontaneously rupture or are broken by accident, 
and then lymph drains away. Sometimes it simply streams 
away, and I have collected an ounce (30 grams) in one and a half 
minutes from one opening. 

The strain on the patient due to this constant flow is most 
trying, and I have seen a patient in a critical condition from 
sheer exhaustion due to this cause. Such acute cases are for- 
tunately rare, and the main complaint that the patient makes 
is of discomfort due to his clothes being always in a wet con- 

Another point that should be borne in mind is that the dis- 
comfort of a lymph scrotum specially appears during or after 
an attack of lymph fever. Although the fever itself may have 
abated, there remains a recently inflamed lymphatic tract con- 
stituting a field of infection, ready to hand, in the area of scrotal 
lymphorrhagia. Occasionally most troublesome inflammation of 
the whole lymphatic region starts in this way, spreading into 
the superficial inguinal regions and then down the thighs fully 
halfway to the knees. 

Erysipelatoid inflammation and elephantoid fever are common 
concomitants of lymph scrotum; and abscess, either preceding 
or accompanying the affection, is not rare. It must be remem- 
bered that all varieties of lymph scrotum are met with, from 

296 '^h^ Philippine Journal of Science 1921 * 

that in which the affection is just becoming manifest to that 
in which the scrotum is one mass of large vesicles. 

Another very common type is that in which the scrotum is 
partly varicose and party elephantoid. This is particularly likely 
to 'be the case if the patient has suffered much from erysipe- 
latoid inflammation in addition to the lymph scrotum. If, 
on the other hand, he has only suffered from elephantoid fever, 
and to a very limited extent from erysipelatoid inflammation, 
then the scrotum will remain a lymph scrotum to the end. 

It is difficult to follow up these cases in the Far East but, as 
a rule, they take one of two roads : either they become elephan- 
toid, in which case they may develop into large tumors like 
those scrota that are elephantoid from the beginning, or they 
remain varicose. In the latter event they tend, as the patient 
reaches to 50 years or more of age, to become atrophic. Some 
of the patients have had the disease for over thirty years. 

Two cases were kept under observation for four years. In 
one, in spite of recurring attacks of elephantoid fever, the 
scrotum steadily shrunk. This patient was over 55 years of 
age. In the other case there were fewer attacks, but the ery- 
sipelatoid inflammation was more marked, and the scrotum 
double in size and was rapidly becoming an elephantiasis 

Embryo filarise are generally found in the blood in these cases, 
and very often they may be obtained from the lymph flowing 
from the scrotum. On one occasion, as mentioned on a previous 
page, filarial ova were found in the fluid from a scrotum of this 
kind, when repeated examination had failed to discover the 
embryo in the blood. In this case the fluid flowing from the 
varices was straw-colored, and the groin glands were not en- 

The past history of these cases varies greatly. Here is a 
patient who had suffered from elephantoid fever for at least 
fifteen years, and whose scrotum had only during the last year 
become a lymph scrotum. Here is another patient who, within 
six months of his first appearance of trouble due to the filaria, 
had developed a very marked condition of lymph scrotum; but 
as a rule a marked condition of lymph scrotum takes about a year 
to develop, and every degree of varicosity is produced. The 
largest varices are generally found near the root of the penis. 
Occasionally the testicles are a little thickened, especially if 
there has been an attack of filiarial orchitis. Also, single or 
double hydrocele may or may not be present. 

19,3 Maxwell: Filariasis in China 297 

Treatment may be divided into that of the acute attacks and 
that of the general condition. In acute attacks, the best place 
for the patient is in bed, with his scrotum supported at a higher 
level on a pillow. This simple proceeding often gives great 
relief. Strict cleanliness must be observed, and the scrotum 
must be treated by either a dry or a wet method. In the former 
it is dried and powdered with any antiseptic powder. A powder 
composed of zinc oxide, or a mixture of this with boracic acid, 
suits well. If a wet dressing is to be used and sometimes it 
suits a case much better than the dry one, lint kept continually 
wet with cold water or lotio plumbi subacetatis may be used. 
Occasionally a little opium added to the lotion is of advantage. 
Each case, however, must be treated on its merits. 

Where the lymphorrhagia was very severe and mostly confined 
to a single spot, I have several times ligatured the lymphatic 
vessel, using a fine needle and horsehair, leaving the ligature in 
place for a day or two. 

The treatment of the general condition raises large questions ; 
for operation, though comparatively easy, involves the great risk 
of setting up trouble in some other part of the lymphatic area. 
Operation has been followed by the commencement of chyluria 
or elephantiasis of the leg. More than this, the wound is very 
large and takes a long time to heal, even if Thiersch grafting be 
employed. Six weeks to two months is not too long to estimate. 

Supposing that operation has been decided upon, what opera- 
tion should be performed? Manson(32) gives the following 
directions : 

The scrotum should be well dragged down by an assistant whilst the 
testes are pushed up out of the way of injury. A finger-knife is then 
passed through the scrotum and in sound tissues, just clear of the testes, 
and the mass excised by cutting backwards and forwards. No diseased 
tissue and hardly any flap should be left. Sufficient covering for the 
testes can be got by dragging on and if necessary dissecting up the skin on 
the thighs, which readily yields and affords ample covering. It is a very 
common but a very great mistake to remove too little. As a rule, the wound, 
if carefully stitched and dressed antiseptically, heals rapidly. 

I am not in favor of this method, as in many cases it will fail 
to remove the whole disease, and the bleeding may be serious. 
I prefer an operation similar to the one undertaken in the follow- 
ing case: 

A semicircular incision was made from side to side across and 
under the root of the penis and well up into each inguinal region. 
This incision was carried through the skin and underlying fascia, 
all vessels being clamped as cut. Two lateral incisions were then 

298 The Philippine Journal of Science 1921 

made from the extremities of this first incision on the lateral 
aspect of the scrotum from base to apex. The flap thus marked 
out was turned down, laying bare both testicles and their cords. 
These were freed and turned forward, the whole of the remainder 
of the scrotum being then cut away. The testicles were sewn 
in position to the wound in the perineum, and the skin from the 
lateral aspects of the thighs was drawn partially over them. 
Even though this was done, there still remained a wide, gaping 
wound which, in spite of Thiersch grafting, took six weeks to 
heal. The final result was very satisfactory, as the attacks 
of elephantoid fever ceased from that moment and had not re- 
curred two and a half years later. 

Many natives have asked for operation, but when they learned 
the time that they would have to stay in the hospital, they backed 
out, saying that the discomfort was not sufficient to make them 
give all that time to the cure. In the slight cases, it is not worth 
their while ; and in the severe ones, there is no chance of obtain- 
ing flaps, a condition which greatly delays convalescence. 

If it is decided not to operate, then the patient should be in- 
structed to keep the parts scrupulously clean, suspended, and 
protected from undue friction between the thighs. 


Fortunately for the subjects of filarial disease, chyluria is by 
no means common. It may begin in several ways. Sometimes 
it begins without warning and is the first indication to the patient 
that he is in any way out of health; but more often there are 
some premonitory symptoms, either in the form of dragging 
pain in the back or aching in the groins or hypogastric region. 
If the patient is at the same time the subject of one of the other 
forms of filarial disease, the onset may be heralded by the trouble 
pertaining to that form. For example, here is a man, the sub- 
ject of lymph scrotum. In his case the first onset of the 
chyluria was associated with an attack of elephantoid fever and 
scrotal lymphorrhagia — and this is only natural ; whatever raises 
the pressure in the already existing lymphatic varix is liable to 
precipitate the onset of chyluria. 

The pathology of the affection is quite clear. Given a lym- 
phatic varix in the wall of the bladder or in the region of the 
pelvis of the kidney, you have only to raise the pressure so that 
the dilated lymphatic ruptures in one or the other situation, and 
you have chyluria. This chyluria may be intermittent or per- 
manent. In the intermittent cases the chyluria only occurs 

19,3 Maxwell: Filariasis in China 299 

under provocation, such as is afforded by an attack of elephantoid 
fever, and in the intervals the urine is perfectly normal. It is 
important to bear this in mind, as the practitioner in a filarial 
region is sure, sooner or later, to meet with patients who complain 
of having passed bloody urine, in whom the urine is normal at 
time of consultation ; in such cases the explanation is that they 
have had an attack of chyluria of sanguineous hue and, on 
questioning these patients carefully, the point may be elicited that 
they had an attack of fever at the time. In what have been called 
the intermittent cases the attack lasts a day or two at the longest. 

The cases that have been called permanent are not truly so, 
for with rare exceptions the chyluria does not continue per- 
manent through life ; but it may last weeks or months at a time, 
then intermit, and at an uncertain and variable time reappear. 

It must be remembered that the disappearance of the milky 
nature of the urine is not proof that the fistula is closed, as the 
urine may still contain proof of the lymphatic connection in the 
presence of albumen, lymph corpuscles, aiid a gelatinous clot. 

Rarely, retention of urine from clot formation is the first sign 
of chyluria; and although it is true that in most cases this re- 
tention passes away spontaneously in the course of a few hours 
by the passage of the offending clots, yet their occurrence gives 
the patient much pain and inconvenience. 

The general health of the subject of this disease is, on the 
whole, fairly good ; but if the drain of chyle is severe, the patient 
may become ansemic, depressed, and melancholic. 

The patient generally notices the color of the urine himself; 
but if its milky nature is not marked, the doctor may be the first 
to discover it in the course of a routine examination of his pa- 
tient. The urine may be markedly milky, or pinkish, or red, 
and this color may vary widely in the course of a single day. 
It may pass from milkiness to limpidity in a couple of hours. 
The nature of the food taken also makes a great difference in 
the appearance of the urine, as will be noted in more detail when 
we come to speak of treatment. 

What are the characters of chylous urine? As to odor, this 
may be best described as heavy and urinous. This urine 
decomposes very rapidly in hot climates. Whether decomposi- 
tion would take place with the same rapidity in a cold climate 
is open to doubt, but I have no data on this point. It is said that 
the quantity of urine passed is in excess of the normal ; but the 
quantity of urine passed, especially in a hot climate, depends so 

300 The Philippine Journal of Science 1921 

much on the amount of fluid taken, and the amount of the same 
lost by sweating, that it is impossible to make a definite state- 
ment on this point. The reaction and specific gravity also vary 
within wide limits. 

If chylous urine is passed into a urine glass, in many cases 
the whole becomes coagulated within a very short time. This 
coagulum slowly contracts, becoming redder, denser, and more 
fibrous. The clot at first floats in a milky fluid, which in due time 
separates into three layers, consisting of a whitish pellicle on 
the surface, in the middle a thick stratum containing the 
coagulum in the meshes of which embryo filarise can be usually 
found, and at the bottom a little reddish sediment which some- 
times contains a few small clots of blood. 

The main part of the fatty matter is contained in the middle 
layer in a granular condition. Microscopical examination shows 
the lowest stratum to contain lymphocytes, cells like red blood 
cells (and in some cases probably true red blood corpuscles), 
urinary epithelium, crystals, and a few filarise. Of course, in 
some cases filarial embryos are absent, as in cases where, the 
parent worm having died, the lymphatic varix still persists. 
One good way of finding them is to centrifugalize the urine 
as soon as passed. This way has acted well in the few cases I 
have personally had to treat. 

Ether shaken up with the urine removes the fat, and boiling 
the urine brings down the albumen that it contains. 

Usually the urine is most chylous toward evening, and Man- 
son (33) has clearly shown that the presence of the filar ia in the 
urine is not regulated by the law of periodicity governing its 
presence in the blood. 

Treatment is eminently unsatisfactory. It cannot be said that 
we have any drug that has been proved to have any effect what- 
ever on the disease; nor is it clear what some of the drugs 
used are intended to do. If they are intended to kill the parent 
worm, their use may be very dangerous. Chyluria is not a dis- 
ease that can be called dangerous to life, whereas a dead parent 
worm at the back of the abdomen is a very different matter. 
Personally I would rather keep the chyluria, if I were so unfor- 
nate as to acquire it. If it is intended to diminish the pressure 
in the lymphatic area, it. is difficult to see how these drugs are 
going to accomplish the desired end. Those that have been 
specially recommended in the treatment of this disease are the 
following: Gallic and benzoic acids, thymol, beta-naphthol, methy- 

19,3 Maxwell: Filariasis in China 301 

lene blue, glycerin, perchloride of iron, chromic acid, quinine, 
sodium salicylate, and ichthyol. 

The best results in the treatment of the disease are to be ob- 
tained from general treatment. The patient should be placed in 
bed, and absolute rest should be prescribed; as the patient does 
not usually feel very ill, this absolute rest is by no means easy to 
obtain. The pelvis should be elevated, and the amount of food 
and drink should be restricted. The bowels should be well 
cleared out. 

If a minimum of fatty and fluid food is administered, the 
milky character of the urine may quickly pass away. This 
does not necessarily mean that the fistula is closed, as I have 
previously stated; but very often, by the above treatment the 
fistula can be really got to close and remain closed, sometimes 
for many months. 

The subjects of this affection should beware of excessive or 
very violent exercise, as it is probable that an attack may be 
precipitated in this way (see next section). Special care must 
be taken by women during pregnancy, as the extra strain of 
labor is not at all unlikely to bring on an attack in susceptible 

Clots of large size form occasionally in the bladder and give 
much trouble in the way of retention. When this occurs they 
may be easily broken up with the aid of a metal catheter. 


Chylous dropsy of the peritoneum is extremely rare. Win- 
kel(45) records a case that was probably of this nature. The 
first serious symptom was the development of ascites. The fluid 
drawn off by tapping was of a milky character and contained 
what were probably filarise. Afterward the patient developed 
swelling of the left leg, which gradually subsided. The ascites 
returning, she was again tapped, and fluid of the same character 
was drawn off. Two days later she died suddenly. 

One case has come under my care. The patient was a man 
of middle age, who had suffered from elephantoid fever. After 
a wine party he went outside and "saw a demon." He was 
greatly frightened and ran all the way home, a distance of three- 
quarters of a mile (1.2 kilometers) . He felt uneasy in the abdo- 
men on reaching home, and by morning it was swollen. A week 
later he came to hospital. His abdomen was moderately swollen 
by free fluid. On tapping, milky fluid was drawn off, but no 
filaria was found in it. The patient refused treatment and left 

302 ^^^ Philippine Journal of Science 1921 

the hospital at once. He did not seem to be acutely ill. The 
fluid coagulated on standing, but not completely. 

It is remarkable that this, form of the disease is not more 
frequent. It is possible that small leakages take place more 
frequently than is known and that the peritoneum is able to 
dispose of such leakages without their actually causing ascites. 


Chylous dropsy of the tunica vaginalis is by no means as rare 
as chylous dropsy of the peritoneum. Presenting the characters 
of an ordinary hydrocele, with the one exception of translucency, 
it is very often undiscovered until tapped, when milky fluid 
flows through the cannula. This milky fluid generally contains 
many filarise, quite out of proportion to what one might expect. 
Parent worms have never been found in one of these lympho- 
celes, which rarely attain great size. The fluid drawn off from 
one of these swellings does not always coagulate. 

Lymphocele is usually associated with some other manifesta- 
tion of filarial disease, such as lymph scrotum, varicose groin 
glands, or elephantoid fever. 

It should be treated as an ordinary hydrocele. If not large, 
it may be let alone or treated by excision of the sac. This is a 
much safer way than injection. Certainly, if a case of this 
kind were to go wrong after injection, it would be much more 
serious than in the case of an ordinary hydrocele, owing to the 
varicosity of the lymphatics. This raises the whole subject of 
the relation of ordinary hydrocele to filarial disease. Man- 
son, (34) years ago, suggested that the great frequency of hydro- 
cele in tropical countries might be due to the presence of filarial 
disease, and James (16) advances the same view and gives a few 
points bearing on the subject. The matter is not yet cleared up, 
and from what I have observed I doubt that filarial infection is 
the cause of more than a small percentage of cases. Out of the 
last series of Charles's (7) elephantiasis operation cases, eighty 
in number, fifty-nine suffered from hydrocele. Of my own 
operation cases, every one suffered from hydrocele, small or 
large, so that I may justly say that in elephantiasis cases hydro- 
cele is the rule and not the exception. Of course, it may be 
alleged that this is a secondary matter, and primarily not due 
to the filarial disease, but to the elephantoid change in the tissues. 

Excluding these cases, certainly not above 10 per cent of the 
hydroceles in the coast belt with which I have had to deal by 
tapping or other operation have been the subjects of filarial 

19.3 Maxwell: Filariasis in China 303 

infection. When compared with the figures given for general 
incidence, it must be admitted that the filarial disease, barring the 
elephantoid form, does not seemi to be especially prone to give 
rise to hydrocele. 


Chylous diarrhoea is extremely rare, and I have never seen a 
case. As the lymphatic varix must extend in some cases into 
the wall of the bowel, it is by no means an unexplainable trouble. 
As such cases are only a curiosity, they need not detain us fur- 

We have now to discuss the forms characterized by the pro- 
duction of more or less solid oedema in the parts affected, in 
addition to the lymphatic dilatation that characterizes all the 
affections that I have grouped under this section. 

Not so very long ago it was necessary to enter into elaborate 
reasons which would justify the placing of the diseases char- 
acterized by solid oedema among the filarial diseases at all. 
Happily, with the advance of knowledge, this is no longer 
necessary ; still, it is well to bear in mind the main reasons that 
justify this classification. It is true that not all cases of ele- 
phantiasis are connected with filarial disease and that, while 
the most common cause, it is not the only one. In the East and 
in Europe cases occur that have nothing to do with filarial 
disease and yet are as pronounced cases of elephantiasis as one 
could wish to meet; but, on the whole, it may be fairly stated 
that the elephantiasis of China is due to filarial disease, for the 
following reasons : 

a. The distribution of filarial disease and elephantiasis are the same. 

Generally speaking, where there is no filaria, there is practically 
no elephantiasis, and vice versa. After a period of work in the 
coast belt of Fukien, I moved inland to a place well outside the belt. 
Here filariasis was absent, and the few cases of elephantiasis of 
the leg that came to me were either frankly due to other causes or 
were cases that had originated in the coast belt. 

b. Every gradation between lymph scrotum — which is clearly a filarial 

disease — and elephantiasis scroti can be observed. 

c. Both lymph scrotum and elephantiasis scroti are lymphatic varices. 

d. The type of fever observed in undoubted filarial cases and elephan- 

tiasis cases is precisely the same. 

e. The removal of a lymph scrotum or varicose groin glands — both 

manifestly filarial diseases — ^may be followed by the development 
of a form of elephantoid disease, either elephantiasis of scrotum 
or leg. 
/. Acknowledged filarial disease and elephantiasis may and often do 
occur in the same subject. 

304 The Philippine Journal of Science 1921 

Before taking up these diseases in detail, it will be well to 
discuss their pathology; and it must be freely conceded that in 
the majority of cases filar ise will not be found in the peripheral 
blood. Probably this is due to one of two causes : either to the 
death of the parasite or to the blocking of the lymphatics drain- 
ing any affected area, so that the filarise cannot make their way 
into the blood. The solid oedema characterizing the disease is 
brought about not by the mere obstruction of the lymphatics, but 
by a combination of this with inflammation set up in the blocked 
area. But what causes the blockage? First there is the action 
of the parent worm, which, from injury or some other cause, 
may abort. The ova then either block the lymphatics or obstruct 
the glands of the area in question, producing a lymph stasis. 
In this area, already in a condition of lymph stasis, inflammation 
takes place, and an abscess may form which destroys the lym- 
phatics in its vicinity. Probably this plays a much greater 
part in the production of elephantiasis scroti than is supposed, 
and it is the common thing in Fukien to be able to elicit from 
elephantiasis patients a history of abscess in the scrotum, often 
the first sign of trouble in that part. To this are superadded 
frequent attacks of erysipelatoid inflammation, each attack aiding 
in the production of the permanently thickened condition of 
the tissues. In the case of the scrotum it will be observed that 
the tumor increases in size very slowly, up to a certain point, 
and then increases at a much more rapid rate. This phenom- 
enon is due to the accident of position. As long as the scrotal 
tissues are able to support the weight, the tumor remains braced 
up, so to speak. As soon as the tumor reaches a certain size, 
the tissues yield and relax. The weight of the tumor then comes 
into play, and all the lymphatics passing over the upper edge 
of the pubis are further obstructed. 

In the case of the leg, the dependent position of the part 
plays no small share in the production of the enlargement, and 
in an operation case of my own the forced rest involved by the 
operation for elephantiasis scroti sufficed to reduce the leg tem- 
porarily to a third of its former size. When one cuts into an 
elephantoid mass such as is found in elephantiasis scroti, after 
traversing the greatly hypertrophied skin, the subdermal tissues 
are found to be also much thickened, and the dilated lymphatics 
are easily made out; below this again is a mass of blubbery 
tissue containing much fluid, and this is easily traversed by the 

19,3 Maxwell: Filariasis in China 305 

The rarity of elephantiasis of the upper parts, of the body is 
explained by the rare lodgment of the parasite in these regions 
and by the free anastomosis of the lymphatics in these parts. 

We come now to a discussion of erysipelatoid inflammation, 
which plays a large part in the formation of a fully developed 
case of elephantiasis of the leg. There are two views generally 
held. Many hold that the essential cause is a low form of strep- 
tococcic infection superadded on a limb that is already the subject 
of lymph stasis. Those who hold this view are unwilling to 
concede that the filaria plays any active role at all, and some 
deny even the passive role referred to above. (37) 

Others, including myself, are convinced that elephantiasis of 
the leg is infinitely more common in filaria-infected regions 
than in those free from the infection, and are not convinced 
that the filaria plays a merely passive role in the production of 
lymph stasis and the subsequent elephantoid development, hold- 
ing that the results of bacteriological examination of the tissues 
are not conclusive and that the inflammatory attacks may be 
due to a poison generated in the body itself. Why, for ins.tance, 
should a case of lymphatic stasis due to sloughing of the glands 
in a case of plague be practically free from the constant inflam- 
matory attacks common in a filarial case, and why should a 
lymphatic fistula ward off these inflammatory attacks in a fila- 
rial case without preventing the slow increase in the size of the 
parts below the fistula? More investigation is badly needed in 
these cases, both of the exact nature of the obstruction and of 
the nature of these erysipelatoid attacks, before the question can 
be deemed to be satisfactorily settled. 


Elephantiasis scroti generally begins with attacks of ele- 
phantoid fever combined with erysipelatous inflammation of the 
scrotum. The part becomes swollen and oedematous, and hydro- 
cele not infrequently makes its appearance. The acute attack 
subsides, but leaves behind a legacy in the shape of some per- 
manent thickening and enlargement. This process recurs and 
is varied perhaps by the occurrence of suppuration in the 
scrotum. In whatever way started, the thickening tends pro- 
gressively to increase, and occasionally there are only a few 
attacks of fever throughout the whole disease. 

The disease is not uncommon in the coast belt, but as a rule 
the tumors are not of large size; and so little trouble do the 

181052 1 

306 "J^^^ Philippine Journal of Science 1921 

smaller tumors give that the patients are usually not willing to 
submit to operation, unless the condition is such as to prevent 
sexual relations. 

As to the tumor itself, the scrotum is greatly increased in 
bulk, and the skin is thickened and rugose. Occasionally this 
rugosity is so marked as to make the surface appear to be 
covered as if by a crop of warts. The hair is scanty, dry, and 
coarse, and the mouths of the follicles stand out very distinctly. 
The skin is also firmly bound dovm to the subjacent tissue, and 
as a rule, when fully developed, the scrotum does not pit on pres- 
sure. The sweat and sebaceous glands, as a rule, are atrophied. 
It is generally very difficult to say precisely where the healthy 
skin ends and the unhealthy skin begins. The blood vessels of 
the scrotum are enlarged and slightly thickened, but this thicken- 
ing is as nothing when compared with the thickening to be seen in 
the lymphatics. Some of thefee at operation stand out as large, 
open-mouthed vessels. A hydrocele is often found with walls 
thicker than usual. If a hernia of old standing be present, the 
cord on that side may be so spread out over the hernial sac as 
to make it a matter of difficulty to recognize the former. 

The weight of these scrotal tumors varies very much. All 
sizes may be met, from the tumor weighing only a few ounces 
(grams) to the largest recorded of 224 pounds (102 kilo- 
grams). (35) The largest that I have operated on weighed, 
after removal, 71 pounds (32 kilograms) . From 5 to 50 pounds 
(2.2 to 22.6 kilograms) is the commonest weight. 

These scrotal tumors may be classified in two ways ; namely, 
according to the clinical characters of the swelling or according 
to the clinical characters of its neck. 

According to the first classification, we meet with four forms : 

a. A hard and solid type with no hydroceles or only very small ones. 
h. One with large hydroceles and a fair amount of elephanto-id tissue, 
involving more than the skin. 

c. Elephantoid disease of the skin only, around a large hernia or 


d. Elephantiasis of the penile skin, the scrotum remaining practically 


According to the second method of classification, the tumors 
are divided thus : 

a. Narrow-necked tumors. 
h. Broad-necked tumors. 

This may seem an arbitrary and unnecessary distinction, 
but its practical use is great; for the one kind of tumor, the 

19,3 Maxwell: Filariasis in China 307 

narrow-necked, is an easy one on which to operate, whereas 
the broad-necked tumor requires considerable skill. 

In considering the treatment to be advised for a patient suf- 
fering from elephantiasis scroti, it must be remembered that 
in many instances these scrotal tumors never reach a large size 
and, provided they are giving little trouble, are quite as well let 
alone. If they are growing or if they interfere with the per- 
formance of the sexual act, removal should be undertaken. In 
calculating statistics of mortality, the cases of small tumors 
should be placed by themselves, for the mortality from mere 
decortication of the scrotum, in which the portion removed 
weighs only a few ounces (grams) ought with reasonable care 
to be practically nil; whereas the removal of a large scrotal 
tumor, which in many instances cannot be properly disinfected, 
is by no means devoid of risk. 

In the case of the larger tumors, including those with compli- 
cations such as hernia, which sometimes are very serious, the 
mortality in competent hands should not be above 2 per cent. 

We will now discuss the most suitable form of operation. 
Careful local preparation is essential. This is carried out as fol- 
lows: The parts are shaved, and then soap and water are lib- 
erally applied with the aid of a nailbrush. As dirt is generally 
much in evidence, this should be done several days in succession. 
Any open sores and sinuses have to be dealt with. On large 
tumors these are of common occurrence, the sinuses being the 
remnants of old abscesses and sores, the result of the itching of 
eczema or ringworm which has led to scratching. If eczema or 
ringworm is at all severe, the operation should be delayed until 
this condition is improved. 

On the night before operation a final scrubbing is given, fol- 
lowed by scrubbing with turpentine to remove fatty matter, 
and by careful washing with 1 to 500 biniodide of mercury 
lotion in weak spirit. This again is washed off with biniodide 
of mercury lotion, 1 to 4,000, and a dressing of gauze soaked 
in the same solution is applied, covered with protective, and kept 
in place with a bandage. 

Another method is to allow the parts to get thoroughly dry 
after the preliminary preparation and then to employ either an 
iodine solution or one of the newer dyes, such as methyl violet 
and brilliant green, for the final disinfection before operation. 
J^ the operation the greatest care shouldibe taken, by means of 
sterilized towels, to isolate the operation area and the instru- 
ments, and ligatures should be most carefully prepared. 

308 ^/^^ Philippine Journal of Science 1921 

As to the position of the patient, the lithotomy position main- 
tained by a Clover's crutch or leg-rests should be used. This is 
by far the most convenient. When operating on a patient the 
subject of hernia, the hernia is best dealt with first by the usual 
incision, with the patient lying on his back. Thereafter the 
patient is put into the lithotomy position for the removal of the 
scrotum. During the general preparation of the patient, the 
bowels should be well cleared out with an effective purgative, 
best given on the second evening before operation, so as to 
ensure the patient's getting an undisturbed night before the 
event. Any cough should be treated, and the heart and urine 
examined in the routine way. It is difficult to give definite 
advice about operation on opium smokers, such a habit adding 
considerably to the risk of any large operation. Those who 
smoke heavily should be compelled to break it off before oper- 
ation; otherwise the operator will have no end of trouble with 
them afterward. Patients who are the subjects of malarial 
cachexia should be guarded from unnecessary danger by pre- 
vious suitable treatment. 

I never kept the patient in bed before the operation, believing 
that the gain in the reduction in size of the tumor is little com- 
pared with the depression arising from the concentration of 
the patient's mind on his complaint and on the operation, which 
he naturally dreads. So he is allowed to do as he likes until 
the night before the operation. 

Anatomy of the parts in question, — It must be borne in mind 
that all the vessels are greatly increased in size, and for this 
reason vessels that are otherwise quite unimportant may become 
serious sources of hsemorrhage. 

The important vessels are the following : 

Superficial vessels: Superior external pudic vessels. 

Deep or inferior external pudic vessels. 

Sciatic vessels, muscular branches. 

Superficial perineal vessels. 

Cremasteric vessels. 

Dorsal vessels of the penis. 

Artery of the frenum. 

Inferior haemorrhoidal vessels, if the skin has to be removed to near 

the region of the anus. 
Deep vessels: Branches of the internal pudic vessel. 

These should be looked for and found in their proper 
situations. The dorsal vein of the penis should not be cut. 

19.3 Maxwell: Filariasis in China 309 

The nerves and their positions may be entirely disregarded. 
Nerves, however, should be cut short, and not left hanging in 
the large wound. 

The lymphatics, also, may be disregarded; but, if large and 
gaping, they should be tied. 

So much for the anatomy of the part. Now we must dis- 
cuss a most important question. Is a tourniquet to be used 
or not? 

Use of a tourniquet. — Many urge the use of a tourniquet 
in order to save loss of blood, but the oozing in such cases after 
operation is frequently severe and troublesome. On the other 
hand, if a tourniquet is not used, the primary loss of blood may 
be severe. On the whole, I think the best advice is that if 
the surgeon is afraid of the primary haemorrhage he should use 
a tourniquet, but if he is familiar with the anatomy of the 
area and is a fairly rapid operator, he will do better without. 
In the earlier cases I always used the tourniquet, but in later 
cases I did not; and since discontinuing its use I have had none 
of the unpleasant after-bleeding that was experienced in one 
or two of the earlier cases. 

Another disadvantage in the tourniquet as applied by Mc- 
Leod's(23) or Hanson's method is that there is a constricting 
band around the abdomen, a thing that is objectionable, as it 
probably increases the danger of the anaesthetic. If a tourni- 
quet is to be used, it is best applied in the following way : 

Across the neck of the tumor is laid a long, sterile calico 
bandage from the right arm to the left foot. Behind, the same 
thing is done from the left arm to the right foot. The tumor 
and the two lower ends of bandage are held up and a piece of 
quarter inch drainage tube is carried around the neck and the 
bandages in a circular manner and tied securely. The lower 
end of the bandage in front is then turned upward toward the 
left arm, and the lower end of the bandage behind is turned up 
toward the right arm. The two ends toward the left arm 
and the two ends toward the right arm are held by an assist- 
ant on each side and so maintain the tourniquet in position. 
The tourniquet and bandages are, of course, carefully sterilized. 
This method obviates the necessity of having a tight band 
around the abdomen. 

Incisions, — Hard and fast rules should not be laid down as 
applying to every case. For instance, in one case in which the 

310 ^he Philippine Journal of Science 1021 

skin of the penis was normal, the whole of it was left and after 
two years the skin was still perfectly sound. 

Again it used to be urged that all diseased skin should be 
removed. If, as sometimes happens, the skin of the anterior 
abdominal wall is extensively involved, to remove the whole 
skin of this region is to make a huge wound without any real 
advantage to the patient. 

Again a man of 60 years should not be treated like a man 
of from 25 to 30 years. In the latter case, to leave unhealthy 
skin is to court recurrence; in the former, it is most unlikely 
that there will be anything of the kind, and the attention of the 
surgeon should be rather drawn in the case of the older man 
to the provision of flaps, so as to insure a quick recovery. 

The same discretion must be shown in dealing with the tes- 
ticles. In a young man these must be certainly preserved, but 
in an old man, if atrophied, and their removal will facilitate 
healing, one or both may be removed. In a case complicated 
with hernia, it is good practice to remove the testicle on the 
side of the hernia. One can then be sure of closing the abdo- 
minal wall securely, and in this case there is the additional 
advantage that the operation is thereby shortened. 

Speaking generally, I use two main forms of incision, one when 
using a tourniquet and another when operating without it. 
When not using a tourniquet, the incisions are made as follows : 

The first is in the median line from the region over the pubis 
to the urinary orifice. Generally, as the tumor grows, the pre- 
puce is inverted, and the glans may be several inches (centi- 
meters) from the external orifice. A finger of the left hand is 
inserted into this orifice, and the incision is rapidly deepened. 
It is very easy to tell when one has exposed the penis. Using 
a pair of scissors, and with the aid of one's finger as a director, 
the prepuce is opened along the dorsum as far as the glans. All 
smegma is carefully cleaned away, and the lining membrane of 
the prepuce is dissected off and left attached to the glans penis. 

The penis is now to be thoroughly separated from its sur 
roundings from the suspensory ligament outward, taking care, 
below, not to cut the artery of the f rsenum. If it is wounded, it 
is better to ligature it at once. Above, care must be taken of 
the vessels on the dorsum of the penis, the vein being always 
very large. If due care is taken, it ought not to be wounded. 
The penis, having been separated, should be wrapped in gauze 
and turned upward. 

19,3 Maxwell: Filariasis in China 311 

The tumor is now pulled well over to the left, and beginning 
over the right external abdominal ring, a lateral flap is marked 
out, the incision ending in the middle line in front of the anus. 
The upper portion of this incision is now deepened, the vessels 
being picked up as cut. The external pudics will both be secured 
in this way. The skin left in this flap should be soft and plia- 
ble, and no oedematous, subcutaneous tissue should be left if it 
can possibly be helped. This incision and proceeding are re- 
peated in the same manner on the other side of the tumor. 

The upper ends of these incisions are now joined by a straight 
transverse incision, which crosses the upper end of that which 
has been used for separating out the penis. 

A free incision is now made over one of the testicles, begin- 
ning from one of the lateral incisions near the external abdominal 
ring. This is rapidly deepened till soft, oedematous tissue is 
reached, when the knife is laid aside and the fingers are inserted. 
With their aid, the oedematous tissue, which runs with fluid, is 
torn open, and the testicle and cord are isolated. It is better 
to isolate the testicle flrst, as it is more easily found; the cord 
is not infrequently much expanded. The gubemaculum testis is 
in these cases very strong, and must be cut away with scissors or 
knife. The testicle is frequently compressed and deformed, and 
the cord is much lengthened, but this lengthened condition of 
cord soon corrects itself and needs no special treatment. If the 
testicle is hopelessly atrophied, then it is as well removed. 

Charles (8) speaks of congestion of the head of the epididymis, 
or fibrous induration of the same, in cases attended with 
hydrocele. I have looked for this and have not been able to 
satisfy myself on the point. 

The testicle having been freed, if there is a hydrocele, it 
should be cut open and the sac cut away, except a strip, which 
is left attached to the cord. The same procedure having been 
repeated on the other side, the testicles with their cords are 
then wrapped in gauze and turned up on the abdominal wall. 

The incisions already made for the flaps are now to be rapidly 
deepened, and the whole mass is cut away from above, all vessels 
being clamped as cut. 

Some operators advise arresting haemorrhage by the clamp 
torsion method. I always ligature the vessels, using fine catgut. 
It is better not to employ silk for these large tumors as, owing to 
the impossibility of complete disinfection of the skin, infection of 
the wound is by no means unknown, and in such case the silk 

312 The Philippine Journal of Science 1921 

ligatures may give trouble. The number of vessels requiring 
ligatures is often very great, from forty to sixty ligatures being 
sometimes needed. 

All bleeding having been arrested and clot sponged av^ay, the 
testicles should be fixed in place by a stitch or tv^o to the 
perineum. If the cord is very long, this too is secured in its 
proper situation by the same method. The flaps are nov^ dravm 
over the testicles from each side and are fastened together by 
stitches of silkv^orm gut. If they do not come easily together, 
this may be facilitated by freeing the skin flaps at their 
attachment to the thigh, using knife and finger for the purpose. 

A gauze drain should be inserted from the region of the base 
of the penis, bringing the end out at the lov^er extremity of 
the v^ound. This drain is removed in from twelve to tv^enty- 
four hours from the time of operation. 

The penis is now drawn well forward, and the skin of the 
flaps is stitched to its base, the sutures going deeper than the 
cellular tissue and getting hold of the flbrous coat beneath. 
The lining membrane of the prepuce is also turned back and 
is used to cover a portion of the penis. The rest of the wound 
is sevm together with silkworm-gut sutures, and the operation 
is complete. 

When using a tourniquet, it is well to commence with the 
median incision, then separate out the two testicles through two 
incisions down the front of the scrotum and, lastly, make the 
flaps and join the incisions. 

Dressings. — Sterile gauze soaked in biniodide of mercury 
lotion, 1 to 4,000, is a good dressing. This is wrung out and is 
allowed to fall and arrange itself on either side of the penis, 
which is bandaged separately with gauze of the same kind. 
Oiled silk is extremely difficult to keep in good condition in the 
East. The dressing must be large, pass out on to both thighs, 
and fit well around the root of the penis. 

The bandaging is an important and difficult matter. A band- 
aging block is an advantage. Calico is too stiff a material for 
the bandage ; it should be made of a softer substance. What is 
called a water bandage does admirably. It must be put on in 
the form of a double spica, crossing and recrossing the perineum, 
and must be firmly fastened at the dangerous angles with safety 
pins. Especial care should be taken of the penis and the part 
just in front of the anus, which are the most liable spots for 
septic infection. 

i9» 3 Maxwell: Filariasis in China 313 

After treatment — The patient must be placed in bed, and his 
knees should be tied together. There is no need to keep him 
absolutely at rest on his back; he may be allowed to lie in any 
way most comfortable. For the first twelve hours he should 
have nothing but a little very hot water to sip ; rice water may 
be given for the next twenty-four hours, and after that whatever 
food he likes, within reason. The English method of after 
operative feeding does not suit the Chinese, and as a rule they 
flatly refuse such things as beef tea. Under the treatment given 
above they do very well. Usually they do not suffer severely 
from shock, but it is not true to say that this can be excluded, 
for occasionally severe shock may manifest itself; this shock 
is to be treated in exactly the same way as in England by 
strychnine, pituitrin, warmth, stimulants, posture, and saline 
infusion if necessary. 

As has been previously said, the drain of gauze is to be re- 
moved from the lower part of the wound at the end of from 
twelve to twenty-four hours. If the flaps^ have been sufficient 
to cover everything, and all goes well, the wound need not be 
touched for seven days, when it will be found to be healed. The 
gauze about the penis should be removed by careful soaking, and 
the surface should be grafted, if need be, by Thiersch's method. 
Great care must be taken to avoid the binding down of the penis 
by cicatricial bands. The bowels should be opened by a laxative 
on the third day. Under favorable circumstances, healing 
should be complete within the month. Complications that may 
attend the convalescence are discussed in the following para- 

Retention of urine. — This generally occurs at once, if it is 
going to occur at all. It is best treated by the passing of a soft 
rubber catheter, and after the first twelve hours the patient 
may be allowed to stand up to pass water. Occasionally, if the 
wound becomes septic, there may be a little difficulty, but it 
rarely needs instrumental relief. 

Recurrent hemorrhage, — This is a most troublesome complica- 
tion when it does occur, and it is interesting to note that it hardly 
ever occurs except in patients in whose cases the tourniquet 
has been used. On one occasion I had to get up in the middle 
of the night, open the wound, and tie no fewer than twelve bleed- 
ing points. It is true, however, that thiS man had been un- 
usually restless. In another case a haematoma that had formed 
over the upper part of the right cord and was becoming septic 

314 T'T^e Philippine Journal of Science 1921 

had to be evacuated. True secondary haemorrhage I have never 
seen in these cases. 

(Edema of the glans penis. — This is very liable to occur if care 
is not taken in bandaging the organ ; but, when it does, the band- 
age must be taken off and reapplied. 

Troublesome erections. — These sometimes occur, and they are 
a source of annoyance to the patient. If ice can be obtained — 
an impossibility in Fukien except in the coast ports — ^then an ice 
bag may be applied to the perineum. One useful plan is to in- 
sist on the patient emptying his bladder before sleeping, also 
v^henever he wakes during the night, and to forbid him to sleep 
on his back. My experience of bromide and similar powders is 
distinctly disappointing. 

Irritation of the skin. — This sometimes occurs as a result of 
the antiseptics used, helped also by the amount of scrubbing that 
the skin has had to undergo in order to make it anything like 
clean. It must be treated by powdering the surface with zinc 
oxide and changing the dressings for others that are less 

Fever. — There are four causes of fever for which one must 
always be on the lookout. The first is malaria. In the southern 
region, where practically every one carries the Plasmodium 
about with him, it is wise to give the patient a course of quinine 
before the operation. If this has not been done, then it is not 
at all unlikely that the shock of the operation will bring on an 
attack, in which case Plasmodium malarise will be found in the 
blood. It is generally an atypical form of fever. 

The second cause of fever is filarial. One of my cases had such 
an attack on the second day after operation. In this case the 
skin of the penis had been left untouched, and the erysipelatoid 
inflammation ^ of this part was well marked. Almost the whol6 
wound healed by first intention, and the attack did not appear 
to have delayed convalescence. The patient's blood was swarm- 
ing with filariae at the time. Such an attack should be treated 
on general principles. 

The third cause of fever is sepsis and need not be described, 
as it follows the usual course and presents the usual signs. If 
it occurs, a sharp lookout must be kept for an abscess or a sup- 
purating hsematoma, and free drainage must be provided. 

• This typical erysipelatoid inflammation, if due to a streptococcal in- 
fection from the outside, should have produced much more serious results 
in this case. 

19, 3 Maxwell: Filariasis in China 315 

The fourth cause of fever is pneumonia. The utmost care 
must be taken at the time of operation that the patient does not 
get chilled. If he should be so unfortunate as to get pneumonia, 
the case must be carefully watched and treated on ordinary- 

So much for the complications of convalescence, but there are 
yet to be discussed the complications of operation. 

Hernia. — In the principal writings on elephantiasis scroti, so 
little notice is taken of this complication that one is led to believe 
it does not greatly increase the risk of operation. It is nearer 
the truth to say that a large hernia is the most serious com- 
plication one can meet, for two reasons: first, that it is some- 
times difficult to tell the size and contents of the hernia, owing 
to the thickness and hardness of the skin covering the parts; 
and, secondly, because the dragging of the tumor on the muscles 
of the abdominal wall, combined with the length of time the 
hernia has been down, has diminished the capacity of the abdo- 
men, leaving little room for reduction. 

It may be that I have been extremely unfortunate in the cases 
that I have had, but two of them gave me an anxious time. 
The first was a case in which there was a right inguinal hernia 
that had existed for years and was apparently reducible. 
Radical cure was performed on that side, the right testicle being 
removed and the scrotum amputated. Great difficulty was ex- 
perienced in reducing the bowel, and the abdomen when the 
wound was sewn up was as tight as a drum. This case did well 
after looking unfavorable for the first few days. 

The second case was more unfortunate. The patient was a 
man of 44 years, with elephantiasis scroti which weighed, after 
removal, 18 pounds (8.2 kilograms). He had a large right in- 
guinal hernia, the size of which was hard to estimate accurately, 
owing to the very thick skin that extended well up on the ab- 
dominal wall. I was strongly of the opinion that it was not all 
reducible, and, owing to the risk, was rather unwilling to operate. 
However, the man was unable to work, and he was urgent in 
his request that the operation should be done at all risks ; so it 
was finally undertaken. I found myself confronted with a huge 
hernia, in which some of the bowel had evidently been long out 
of the abdomen. After removing 0.5 pound (0.226 kilogram) 
of omentum, and puncturing the bowel in three places to let out 
gas, it was reduced, but the abdomen was so tight that there 
was danger of the stitches giving way. For three days the 

316 The Philippine Journal of Science 1921 

patient did well, then rapidly collapsed and died. It was not 
typical peritonitis, and the puncture holes into the bowel had 
been carefully closed with Lembert sutures; but as no post- 
mortem was possible, it was impossible to learn the precise cause 
of death. Probably it would have been better to have excised 
the bowel ; but after the experience of these two cases it is clear 
that the complication of hernia may be a serious one, and one 
which should call for great care in prognosis. 

Hyd/rocele. — As has been previously stated, the common thing 
is to find hydrocele, and sometimes it is very large. It does not 
complicate the case in any serious way and is treated by excision 
of the sac in the usual method. The thickness of its wall varies 
greatly, from the almost normal condition to one in which the 
whole sac is much thickened. Occasionally this thickening is 

Besides the ordinary form of hydrocele, there are found what 
are called hydroceles of the cord. When present these are to 
be treated by opening them and cutting away part of the wall. 
They are of little importance. 

Varicocele, — This is not common, but is occasionally met with ; 
In such a case the veins should be tied, and a portion should be 
excised in the usual way. 

Cysts, — These are occasionally met with in the body of the 
tumor. They are generally smooth-walled, irregular spaces, 
and may commence in the rupture of some dilated lymphatics 
or may be the remnants of some old lymph abscess cavity. In a 
case in which I found one, it was directly beneath the scar of an 
old abscess. They are merely curiosities and do not affect 

Complicated cases, — Under this heading may be included a 
case like the following: The patient came to the hospital in a 
miserable condition, with an elephantiasis scroti of some 10 
pounds (4.5 kilograms) in weight, a bad stricture of the 
urethra, a right scrotal hernia, and several urinary fistulas. 
The procedure adopted by me was to operate on the hernia by an 
incision placed well up on the abdominal wall, accomplishing a 
radical cure, and tying and severing the cord and pushing down 
the lower end into the scrotum. A week later the patient was 
put into the lithotomy position, and the scrotum was amputated, 
the right testicle being removed with the mass, and the left one 
shelled out. The stricture was now divided by an external ure- 
throtomy, and a metal, full-sized catheter was tied in. All the 
fistulse were followed up to their origin, and the left testicle was 

19. 3 Maxwell: Filariasis in China 317 

sewed in position and covered by a flap from the thigh. The 
penis, which was little affected, was let alone. The patient 
made an excellent recovery and left the hospital with one tiny 
fistula in the scrotum and was able to pass his water by the 
proper channel. He had been taught before discharge to use 
a full-sized, metal sound. A year later he had completely 

What is the result of operations for elephantiasis scroti? 
It can be definitely stated that the results are very satisfactory. 
As a rule, the disease, if effectively treated, shows no tendency 
to recur. The first patient I treated has had two children since 
the operation. One of these was premature and only survived 
its birth a few hours, but the second was a healthy, full-term 

In another case the patient unfortunately was left with a 
lymphatic fistula at the root of the penis. In this case the 
patient was old (63 years), and the elephantoid disease had 
affected the skin of the abdomen as far as the umbilicus, so that 
part of the operation had to be carried out through diseased 
tissues. Under these conditions it is hardly surprising that a 
fistula should be left. 

These favorable results are corroborated by the experience 
of others. 


Elephantiasis vulvae is a rare disease. It may attack the 
whole of the external genitals or some one of their various parts. 
In the latter case the labia majora or the labia minora or the 
prepuce may be affected. There are two main types to be 
distinguished. In the one case the disease is sessile, corre- 
sponding to the broad-necked form of elephantiasis scroti. In 
the case illustrated (Plate 22, fig. 2), the disease had lasted for 
some ten years. The patient had acquired it in the coast belt 
and when seen was also suffering from malignant disease of the 
base of the bladder. Sometimes it involves the two labise, and 
as a rule the sessile form, if it needs operation at all, can be 
removed by an incision going clear of the growth on both the 
outer and inner side. It must be borne in mind that a form of 
elephantiasis vulvae, having nothing to do with filarial disease, 
is by no means so rare. This is due to venereal infection and 
is generally small in size; its aetiology is not obscure. 

The other type is a rarer one, where, a portion of the vulva 
being affected, this local trouble is accentuated by posture, and 
the whole forms a pendulous mass from the region of one or the 

318 ^^^ Philippine Journal of Science 1921 

other labium, varying in size, and comparable with the narrow- 
necked form of elephantiasis scroti. Plates 23 and 24 show 
cases of this kind well. The patients lived in the coast belt and 
had masses of filarial groin glands, which can be seen in the 
photograph of one case (Plate 24). 

In this form of the disease, operation is simple and consists 
of amputating the tumor through the neck. The only point 
especially important is to note that the vessels are stretched and 
should be picked up and tied before they retract. 

Charles (9) recommends that a catheter be introduced into 
the bladder before the operation and be retained there, in order 
to prevent the wound being douched with urine during or after 
the operation. Of course, the operation must be conducted 
under strictly antiseptic precautions. Recovery is generally 
rapid and satisfactory. 


Elephantiasis of the legs may effect one or both legs and is, 
as a rule, much more marked below the knee, although the lower 
part of the thigh is often affected at the same time. Very rarely 
the thigh is affected before the leg, and I have seen at least one 
good example of this condition. The pathology and history of 
the growth of the legs are much the same as those of the scro- 
tum. Very frequently the glands of both groins are varicose, 
even though the elephantiasis is confined to one leg. 

The muscles of the part are frequently atrophied from pres- 
sure, constriction, and disuse. They are also said to undergo 
fatty degeneration from a like cause. The vessels are enlarged, 
and their coats are thickened. The bones may be thickened and 
have osteophytic growths; more rarely they are atrophied. 
The skin is very coarse, does not sweat at all readily, and ife often 
ulcerated, especially about the ankle, where in a normal case of 
elephantiasis there is a constriction permitting of movement 
of the joint. The leg may attain an enormous size, having been 
found as much as 24 inches (61 centimeters) around the ankle. 
Still it is surprising how active some of these patients remain ; 
for example, a patient, a horse keeper whom I used frequently 
to employ, whose leg was not less than 16 inches (40.5 centi- 
meters) in circumference, and who had been in hospital for 
abscess in the leg and for filarial fever. Yet he has walked 
12 miles (19.3 kilometers) after his ponies without any difficulty, 
and that over rough mountain paths. This for a man of 54 
years of age cannot be considered a poor performance. 


19. 3 Maxwell: Filariasis in China 319 

It must be remembered that an abscess in the leg may be the 
first sign of filarial disease, and it is not infrequent for abscesses 
to form in the already elephantoid leg; these may give much 
trouble, owing to their tendency to form lymph fistulas. In 
like manner they are prone to become the seat of intractable 
ulcers, which, however, act as lymph fistulse and so save their 
owners a number of attacks of elephantoid fever. 

Before treatment, one should determine how much inconve- 
nience the patient really suffers. In many cases it is not much, 
and it is as well in such cases to do nothing. If a leg is badly 
ulcerated in addition, it may be amputated, but this should rarely 
be done. I have had to advise it in only one instance, where the 
limb was deformed and ulcerated. Ligature of the femoral 
artery has been practiced ; but, besides being useless, it may be 
followed by gangrene, and is therefore unjustifiable. Elec- 
trolysis has also been used, but it is of no use whatever. Aseptic 
puncture during the acute attacks of erysipelatoid inflammation 
has been performed with good results, but the benefit is only 
temporary; unless done with great care and antiseptic precau- 
tions, it may be followed by ulceration or a lymph fistula. 

The main treatment is prophylactic, and the part should be 
carefully kept from injuries, such as wounds, insect bites, and 
scratches by thorns- If possible, the patient should give up work 
that involves wading in water, with bare legs exposed to the sun. 
If ulceration is already present, it should be prevented from 
spreading over a large area by antiseptic dressing. 

In some cases a spell in bed with the limb slightly raised and 
the use of an elastic bandage will greatly lessen the size of the 
limb. In other cases the constant use of an elastic bandage 
helps the patient. 

In serious surgical treatment, there are several operations 
that have been tried and abandoned. 

Decortication of the limb with subsequent Thiersch grafting 
has been performed. Although in son^e instances the huge 
wound so formed has been induced to heal, the surface broke 
down again, and intractable ulceration and contracture followed, 
necessitating the removal of the limb. 

Longitudinal strips of skin have been dissected off the fascia, 
but here again the procedure proved of doubtful Value, and 
often it was a matter of great difficulty to get the wound to heal. 

There are two operations, however, that have been attended 
with a great measure of success in nonfilarial cases, and they 
probably will be of value in selected filarial cases. One is the 

320 The Philippine Journal of Science 1921 

Sampson-Handley (39) operation for the establishing of a new 
lymphatic circulation; the second is the Kondoleon(i7) opera- 

The first, so far as I am aware, has proved disappointing in 
cases of this kind. It consists in the insertion of silk threads 
leading from the area of lymph stasis into an area above the seat 
of obstruction. Dr. A. F. Cole, late of Ningpo, performed 
several of these operations on cases of this class. In the ones 
that succeeded there was considerable improvement in the size 
of the limb for the time being, but, so far as I am aware, all 
relapsed later. 

The Kondoleon(i7) operation gives more promise of success, 
but so far few operations of this kind have been performed in 
China. One has been reported by Bell, (4) who speaks favor- 
ably of the result, and several have been performed by Dr. E. J. 
Strick of Amoy, with good results ; but hardly sufficient time has 
elapsed for one to be able to judge of the final ending of the 
cases. The principle of the operation is the placing of the deep 
and the superficial lymphatics in connection by the removal of 
a large strip of the deep fascia. The fullest and best description 
of the operation will be found in two papers by Sistrunk, (41) 
of the Mayo Clinic, Rochester. Some of his results are very 
good, but they are in nonfilarial cases, and the amount of ob- 
struction at the back of the abdomen must have its influence in 
determining the final result. At one time it was hoped that 
the injection of fibrolysin would aid greatly in the treatment of 
cases of this class ; but, so far as I know, it has proved unavail- 
ing in producing permanent improvement. The same must also 
be said of the proposed anastomosis of veins and lymphatics 
about the saphenous opening. 


Elephantiasis of the arms is very rare, and I have never seen 
a case in which I could be quite sure of the diagnosis. I have 
seen cellulitis of the arm, of filarial origin, but the limb was not 
a typical case of elephantiasis, and I have not been able to obtain 
information of a well-marked case in China. 

Treatment is unsatisfactory and resolves itself into bandaging, 
massage, and suitable support. It is possible that the Kondo- 
leon operation might greatly benefit such a case. 


Elephantiasis of the mammae also is very rare, and I have 
never met a case. Probably the smaller incidence of filarial 

19, 3 Maxwell: FUariasis in China 321 

disease among women is the cause. If the growth becomes 
large, the breast should be amputated. These tumors are 
occasionally very large and have been known to descend as far 
as the knees. One on record weighed 21 pounds (9.5 kilo- 
grams). (36) 

Cases of hypertrophy of the breast may reach a huge size and 
are found in China, but so far I have seen none to which filarial 
infection could be assigned as a cause. 


Elephantiasis of limited skin areas are also rare, but not so 
rare as the kind just discussed. I have several times seen flat 
elephantoid areas on the thighs, but have never seen any suffi- 
ciently serious to justify operation. 

Localized pedunculated tumors have been reported from other 
localities. Corney, (H) of Fiji, Daniels, (12) and Silcock (40) 
have described cases of this kind. Daniels met with his cases 
in Fiji and Demerara, and Silcock's patient was an East Indian. 
The tumors are easily removed by operation. 



A few observations have been made on the effect of filarial 
infection on the course of other diseases. Surgical wounds in 
the subjects of filarial disease heal very well, unless they happen 
to be in an elephantoid area and have been allowed to become 
septic, in which case they heal very badly. One case, a man who 
had been clawed by a tiger, who came into my hands very badly 
torn and whose blood was swarming vdth filaria, healed well. 
Ordinary operation wounds heal by first intention quite as well 
as in other subjects. Broken bones also unite without any 
trouble. Operations for cataract follow a normal course. 

Of the medical diseases, malaria and filariasis are not antago- 
nistic, and they have been frequently seen combined in the same 
subject. Enlarged malarial spleen and elephantiasis not infre- 
quently coexist in a patient. 

Typhoid fever is not influenced in its course by filariasis. 
Three times I have seen them coexist, without the fever being 
in the least modified. 

Lung diseases are unfavorably modified by the presence of 
filariasis, especially pneulnonia. Twice I have seen patients, 
the subjects of filarial disease, removed in a dying condition by 
their friends; one case died in the hospital, and two or three 

181052 5 

322 The Philippine Journal of Science 1921 

have had narrow escapes. One cannot dogmatize on the matter, 
but probably it will be found that a large number of these fila- 
riasis cases die of lung trouble in the end. A fair number 
suffer from phthisis, but whether the number is greater among 
them than among the general population, it is impossible to say. 

A certain number of such patients suffer from nephritis, but 
what has just been stated about the proportion of phthisis cases 
holds for these also. 

How is the disease to be dealt with as a whole? 

McNaughton (22) has reported a few cases treated by salvar- 
san, in which it was claimed that the parent worm had been 
killed by the treatment. I am not aware of any special work 
that has been done in this direction in China; but, as has been 
stated previously, the killing of the parent worm, even if it 
can be certainly done, is not altogether devoid of risk to the host. 

The extirpation of the mosquito carrier in southern China is a 
matter of impossibility, but much might be done by a proper 
use of the mosquito net or a screening of the bedroom in which 
the infected person sleeps. But what of the number of patients 
who are infected and do not know it? The whole problem is 
one beset with difficulties, and it must be confessed that the 
mosquito net, with all its drawbacks, is the most practical pre- 
ventive measure at our command at the present time. 


I wish to thank many of my fellow workers in southern China 
for kindly answering queries put to them from time to time as 
well as those who have kindly placed photographs at my disposal. 


(1) AZEMA, M. Tropical Diseases. Cassell, London (1917) 715. 

(2) Bancroft, Sr. Discovery of the adult representative of microscopic 

filarias. Lancet, London 2 (1877) 70, communicated by T. S. Cob- 

(3) Bancroft, T. L. Journ. Trop. Med. London 2 (1899) 91. 

(4) Bell, L. N. Am. Journ. Gyn. & Obstet. 29 (1919) 199. 

(5) Brazilian Physicians. Diseases of Warm Climates. Pentland, Lon- 

don (1893) 781. 

(6) Calvert, W. J. Johns Hopkins Bull. 13 (1902) 133. 

(7) Charles, R. H. Ind. Med. Gaz. 36 (1901) 133. 

(8) Idem. Ind. Med. Gaz. 36 (1901) 133. 

(9) Idem. Ind. Med. Gaz. 36 (1901) 133. 

(10) Coles, A. C. The blood in cases affected with filariaris and bilharzia 

haematobia. Brit. Med. Journ. 1 (1902) 1137. 

(11) CoRNEY, B. *'Sibi and Ceke." Lancet, London 1 (1889) 679. 

(12) DANIELS, C. W. Brit. Guiana Med. Ann. (1896). 

19, 3 Maxwell: Filariasis w China 323 

(13) Demarquay. Gaz. Med. de Paris 18 (1863) 665. 

(14) Button, J. E. Brit. Med. Joum. 2 (1901) 612. 

(15) GULLAND, G. L. The condition of the blood in filariasis. Brit. Med. 

Journ. 1 (1902) 831. 

(16) James, S. P. Ind. Med. Gaz. 36 (1901) 114 (letter). 

(17) KONDOLEON, E. Centralbl. f. Chir. 39 (1912) 1022. 

(18) Leiper, R. T. Thelaziasis in man; a summary of recent reports 

of "Circumlocular Filariasis" in Chinese literature with a note 
on the zoological position of the parasite. Brit. Journ. Ophthal- 
mology 1 (1917) 546. 

(19) Lewis, T. R. 8th, 10th, and 14th Annual Reports of the Sanitary 

Commissioner with the Government of India. 

(20) Low, G. C. A recent (observation on Filaria in Culex. Brit. Med. 

Journ. 1 (1900) 1456. 

(21) Maitland. Ind. Med. Gaz. (1898) October. 

(22) McNaughton, J. G. Joum. Trop. Med. 19 (1916) 249. 

(23) McLeod. Tropical Diseases. Cassell, London (1917) 728. 

(24) Manson, p. Trans. Linn. Soc. London 2d ser. Zoology 11, part 10, 


(25) Idem. Hygiene and Diseases of Warm Climates. Pentland, London 

(1893) 765. 

(26) Idem. Tropical Diseases. Cassell, London (1917) 688. 

(27) Idem. Tropical Diseases. Cassell, London (1917) 710. 

(28) Idem. Tropical Diseases. Cassell, London (1917) 711. 

(29) Idem. Tropical Diseases. Cassell, London (1917) 704. 

(30) Idem. Tropical Diseases. Cassell, London (1917) 715. 

(31) Idem. Tropical Diseases. Cassell, London (1917) 713. 

(32) Idem. Tropical Diseases. Cassell, London (1917) 717. 

(33) Idem. Hygiene and Diseases of Warm Climates. Pentland, London 

(1893) 787. 

(34) Idem. Hygiene and Diseases of Warm Climates. Pentland, London 

(1893) 791. 

(35) Idem. Tropical Diseases. Cassell, London (1917) 725. 

(36) Idem. Tropical Diseases. Cassell, London (1917) 733. 

(37) MATAs, R. Am. Joum. Trop. Dis. & Preventive Med. 1 (1913) 60-85. 

(38) MOTY. Revue de Chimrgie (1892) January. 

(39) Sampson-Handley, W. Lymphangioplasty. Lancet, London 1 (1908) 


(40) SiLCOCK. Ind. Med. Gaz. (1895) July. 

(41) SiSTRUNK, W. E. Surg. Gyn. & Obstetrics 26 (1918) 388-393; Mayo 

Clinic Papers 10 (1918) 983-996. 

(42) iSoNSiNO. Med. Times & Gazette, London (1882). 

(43) Stuckey, E. J., and Houghton, H. S. Circumlocular filariasis. China 

Med. Journ. 31 (1917) 24 and Brit. Journ. Ophthalmology 1 (1917) 

(44) Whyte, G. D. Journ. Trop. Med. 12 (1909) 172. 

(45) Winkel. Hygiene and Diseases of Warm Climates. Pentland, Lon- 

don (1893) 789. 

(46) WucHERER. Gaz. Medica de Bahia (1868) December 5; (1869) Sep- 

tember 30. 


[Plates 1, 2, and 3 are reprinted from Tropical Diseases by the kind permission of Messrs. 
Cassell & Company, Limited; Plate 11 is from a photograph by Dr. J. M. Howie; Plate 16, 
from a photograph by Dr. P. B. Cousland ; Plate 16, from a photograph by Dr. J. H. 
Montgomery; Plate 25, from a photograph by Canton Hospital staff; and Plate 26. 
from a photograph by Dr. E. J. Stride.] 

Plate 1 

Anatomy of Filaria bancrofti embryo, a, sheath; b, central viscus; c, V 
spot; d, tail spot. 

Plate 2 

Fig. 1. Section* of thoracic muscles of mosquito about twelve days after 
it had fed on a filariated patient. 
2. Filariae in head or proboscis of mosquito, a, filariae; b, labium; 
c, labium; d, base of hypopharynx; e, duct of veneno-salivary 
gland; /, cephalic ganglia; g, eye; h, oesophagus; j, pharyngeal 

Plate 3 

Metamorphosis of filaria in mosquito, a, b, c, d, e, /, progressive forms 
of the parasite. 

Plate 4 

Section of a varicose groin gland, showing great dilatation of vessels, 
both lymphatic and blood vessels, with hyperplasia of the gland 
tissue; X 60. 

Plate 5 

Fig. 1. Section of skin just becoming affected with elephantoid changes, 
showing a tendency to round-cell infiltration, with dilatation of 
lymph spaces. X 60. The large space is a defect in the section. 
2. Section of skin from fully developed elephantiasis scroti, showing 
dense fibrous felting with thickening of the epithelial layers 
and corium. X 60. 

Plate 6 

Early elephantiasis scroti, showing the scar of a filarial abscess of the 
scrotum on the right side. 

Plate 7 

Lymph scrotum. In this case the swelling of the groin glands is not 
marked. There is some oedema of the preputial skin, and the 
vesicles on the scrotum are well shown. 

Plate 8 

Lymph scrotum; penis unaffected. Shows well the lax, oedematous look 
that some of these cases present. The dark patch is due to 
iodine, which had been painted over large and aching groin glands. 


326 "^he Philippine Journal of Science 1021 

Plate 9 ^ 

Pig. 1. Elephantiasis of penis; scrotum unaffected. 
2. Cauliflower condition of preputial tissues. 

Plate 10 

Elephantiasis of the penis; scrotum slightly affected. 

Plate 11 

Elephantiasis scroti (narrow-necked), weighing, after removal, 32 kilo- 
grams. Patient an opium smoker. The slit in the center rep- 
resents the urinary orifice, the glans being just beyond the reach 
of my finger when inserted at the opening. 

Plate 12 

Condition of parts three weeks after operation. The testicles being atro- 
phied, and the patient in a weak condition at the time of oper- 
ation, these were removed, thus facilitating healing. 

Plate 13 

Elephantiasis scroti. The warty condition of the penis is well shown. 
The skin of the abdomen was slightly affected as far up as the 
umbilicus. Weighed, after removal, 16.3 kilograms. It will be 
evident how Impossible it is properly to disinfect the skin in a 
case of this kind. 

Plate 14 

Elephantiasis scroti, showing a twisted condition, not infrequently seen, 
affecting the penile tissues. Weight after removal, 3.2 kilograms. 

Plate 15 

Fig. 1. Elephantiasis scroti, broad-necked. 
2. Elephantiasis of the cheeks. 

Plate 16 

Condition of parts thirteen days after operation. Shows well the preputial 
lining turned back to form part of the coating of the penis. This 
patient has had two children since operation. 

Plate 17 

(Edema of the penis, which came on shortly after the operation. This 
was due almost certainly to the contraction of the healing wound. 
It passed away of itself within a year, and the patient recovered 
completely. In the right groin the scar of the hernia incision is 

Plate 18 
Elephantiasis scroti. Leg also affected. 

Plate 19 

Elephantiasis of both legs, the left also showing the troublesome chronic 
ulceration met with in some of these affected legs. The fold 
at the ankle and the shapeless condition of the foot are well shown. 

19,3 Maxwell: Filariasis in China 327 

Plate 20 
Elephantiasis of leg, due to plague. 

Plate 21 
Elephantiasis of leg, due to tertiary syphilis. 

Plate 22 

Fig. 1. Elephantiasis of one leg. 

2. Elephantiasis vulvaB, sessile form. 

Plate 23 
Elephantiasis vulvae, pedunculated. 

Plate 24 
Elephantiasis vulvae, pedunculated. 

Plate 25 
Map of China, showing the distribution of filarial disease. 

text figures 

Fig. 1. Chart of elephantoid fever. 

2. Typical chart; filarial abscess. 

3. Chart of filarial abscess. 

4. Chart of neglected case of filarial abscess. 

Maxwell: J^ilariasis in (Jiiina.J 

[t*HrLlP. JOURN. Sci., 19, 1^0. 1 


a. sheath; b, central viscus; c, V spot; d, tail spot. 

Maxwell: Filariasis in China.] 

[Philip. Journ. Sci., 19, No. 3. 

Fia. 1. Filariae in thoracic muscles of mosquito. 

Fia- 2. Filariae In head or proboscis of mosquito. 

Maxwell: Filariasis in China.] 

[Philip. Journ. Sci., 19, No. 3. 

a, b, c, d, e, f, progressive forms of the parasite. 

Maxwell: Filariasis in China.] 

[Philip. Journ. Sci., 19, No. S. 


Maxwell: F'ilariasis iis China.] 

[t*HILIP. JOURN. SCL, 19, No. 3. 

Fig. 1. Section showing early elephantoid changes in skin. 

FIq. 2. Section of skin from case of fully developed elephantiasis. 

Maxwell: Filariasis in China.] 

[Philip. Journ. Sci., 19, No. 8. 





Maxwell: Filariasis in China.] 

[Philip. Journ. Sci., 19, No. 3. 





Maxwell: Filariasis in China.] 

[Philip. Journ. Sci., 19, No. 8. 


Maxwku.: Filariasis in China.1 

[Philip. Journ. Sci., 19, No. 3. 


Maxwell: Filariasis in China.] 

[Philip. Journ. Sci.. 19, No. 3. 


Maxwell: Filariasis in China.] 

[Philip. Journ. Sci., 19, No. 3. 


Maxwell: Filariasis in China.] 

[Philip. Journ. Sci.. 19, No. 3. 






Maxwell: Filariasis in China.] 

[Philip. Journ. Sci., 19, No. 3. 


Maxwell: Filariasis in China.] 

T [Philip. Journ. Sci., 19, No. 8. 













^^ ^; ^ 









■^^-&, * 




Maxwell: Filariasis in China.] 

[Philip. Journ. Sci., 19, No. 8. 


Maxwesa: Filariasis in China.] 

[Philip. Journ. Scl, 19, No. 8. 


Maxwell: Filariasis in China.] 

[Philip. Journ. Sci.. 19, No. JJ. 


Maxwell: Filariasis in China.] 

[Philip. Journ. Sci., 19, No. 3. 


Maxwell: Filariasis in China. J 

[Philip. Journ. Sci., 19. No. 3, 

Fig. L. Elephantiasis of one leg 

P"ig. 2. Elephanllasis vulvae, sessile form. 
PLATE 22. 

Maxwell: Filariasis in China. J 

[Philip. Journ. Sa., 19, No. 3. 


Maxwell: Filariasis in China.] 

[Philip. Journ. Sa.. 19, No. 3. 


Maxwell: Filartasis in China.] 

[Phiup. Journ. Sci., 19, No. 3, 









By Elmer D. Merrill 
Director and Botanist, Bureau of Sdence, Manila 

Burman's Flora Indica was published in 1768.^ The ap- 
pended "Series zoophytorum Indiae Orientalis" occupies 2, and 
the "Flora Capensis Prodromus" 28 additional pages. The work 
is illustrated by 67 plates with 179 figures, the illustrations for 
the most part being excellent. The present review deals only 
with those species proposed as new in the Flora Indica. 

Burman's Flora Indica is of distinct importance from a histo- 
rical standpoint owing to its early date of publication, as it is 
one of the first works other than those of Linnaeus himself, 
published under the Linnean system of classification and nomen- 

It is apparent that most of the species proposed by Burman 
were based on actual specimens, although some of the binomials 
are manifestly based on descriptions and figures in various pre- 
Linnean works. In most cases where new binomials are pro- 
posed by Burman, the figures, the descriptive data from actual 
specimens, or the actual citations of specimens clearly indicate 
that Burman had specimens even when, as is frequently the 
case, he adds in his synonymy references to pre-Linnean works. 
It is hardly necessary to note that, as with Linnaeus, the pre- 
Linnean references are not always correctly placed. 

Burman was in close correspondence with Linnaeus, having 
visited Upsala in 1760, as noted by Jackson,^ bringing with him 
his father's large collection of Cape plants, and was afterwards 
a frequent correspondent.^ Burman himself notes in the intro- 
duction to his Flora Indica that Linnaeus communicated to him 
his observations regarding various species. Duplicate specimens 

^ Burman, N. L. Flora Indica. | cui accedit | series zoophytorum indi- 
corum, I nee non | Prodromus | Florae Capensis. | IV + 1-242, t. 1-67, -f 
indices 1-15, 1768. 

* Jackson, B. D., Index to the Linnean Herbarium, Proc. Linn. Soc. (1912) 
Suppl. 22. 

•Jackson, B. D., op. cit. 11. 


330 The Philippine Journal of Science 1921 

transmitted to Linnaeus by Burman were often described by 
Linnaeus, either under the same binomial Burman used or under 
another name, some in Mantissa Plantarum 1 (1767), one year 
earlier than Burman's publication, others in Mantissa Planta- 
rum 2 (1771), three years after Burman's work was published. 
Of the 241 binomials proposed by Burman in his Flora Indica, 
about 20 were also published by Linnaeus, as indicated above, 
while in the case of about 25 others Linnaeus published the same 
species under binomials other than those used by Burman, but 
manifestly based on material received from him. 

The Flora Indica consists of brief descriptions of about 1,305 
species of which about 241 are proposed as new. Burman's 
work was apparently intended as a descriptive one, covering 
the species from "the Indies" represented in the material 
secured by his father, being for the most part based on collections 
of Piso, Hermann, Garcin, Breyn, Oldenland, Harthog, Kleinhof , 
Outgaerden, and Pryon. The term India is used in a very broad 
sense, covering the Tropics of both hemispheres, although the 
bulk of the material on which the work was based was from 
the Old World. Of the species included more than 500 are 
definitely indicated as from India; that is, mostly from what is 
now known as India proper, although Burman frequently included 
Javan material under the designation "India." In some cases 
more definite localities are given, about 20 being indicated as 
from Coromandel, 25 from Malabar, and a few from other 
localities in India such as Surat and Bengal. Very many of the 
species are indicated as occurring in the Tropics of both hemi- 
spheres. This distribution in many cases is based perhaps not so 
much on actual specimens examined from the Tropics of both 
hemispheres as on material from the Old World, the pantropic 
distribution being based on pre-Linnean synonyms cited as 
representing the species ; in many cases these pre-Linnean refer- 
ences manifestly are erroneously placed. From Java about 115 
species are enumerated; from Ceylon about 90; from China 
about 50; from Japan about 15; from Persia about 20. There 
are a few references to Brazilian plants, and fairly numerous 
ones to those of Amboina and the Moluccas, the latter being 
largely under those binomials primarily based on references to 
Rumphius's Herbarium Amboinense. Other species are enu- 
merated from Jamaica, from Virginia, from Egypt, from 
Arabia, from Peru, from Canada, from Malacca, from Mexico, 
and a few even from the Cape of Good Hope. 

19.3 Merrill: Burman's Flora Indica 331 

While in most cases the geographic origin of the various speci- 
mens cited is correct, some errors have been detected. Thus 
Polypodium spinulosum Burm. f., indicated as originating in 
Java, is the Australian proteaceous Synaphea polymorpha R. 
Br. = S. spinulosa (Burm. f.) Merr. ; Adianthum truncatum 
Burm. f., indicated as originating in Java, is the Australian 
Acacia decipiens R. Br,; and the type of Lotus persicus Burm. f. 
was apparently from South Africa rather than from Persia. 

Burman proposed but five new generic names, of which three 
are universally recognized as valid, and two have been reduced 
as synonyms. These names are as follows : Clausena,^ Embelia, 
Porania, Usubis (=Allophylus) , and Zaleya i=Trianthema) . 

With the exception of those binomials proposed as new, total- 
ing about 241, all the others in the work are accredited to 
Linnaeus. In a considerable number of cases it is perfectly 
evident that Burman erred in his interpretation of Linnean 
species, and some of his misinterpreted binomials have'by later 
authors been made the basis of new names. Thus Panicum re- 
pens (non Linn.) Burm. f. Fl. Ind. (1768) 26, 1. 11, f. 1, although 
a Panicum, scarcely represents the Linnean species; Poa mala* 
barica (non Linn.) Burm. f. op. cit. 27, t 11, /. 2, is Centotheca 
latifolia (Osb.) Trin., the Linnean species being Panicum m/ila- 
baricum (Linn.) Merr. (P. arnottianum Nees) ; Pavetta indica 
(non Linn.) Burm. f . op. cit. 35, t, 13, f, 3, seems to be an Ixora; 
and Vitex pitinata (non Linn.) Burm. f. op. cit. 138, t. i3 [f. 2] 
is Aglaia odorata Lour., the Linnean species being apparently 
identical with Vitex altissima Linn, f.^ These are but a few 
cases that have been noted by me in which Burman's binomial 
represents other than the species Linnaeus intended; there are 
unquestionably numerous other similar ones. 

The younger Burman's collections, on which the Flora Indica 
was based, are preserved in the Delessert herbarium, now at 
the Botanic Garden, Geneva, Switzerland,^ some specimens from 
him being also preserved in the Munich herbarium. The actual 
types, in many cases, have been examined by subsequent authors 
who were engaged in monographic work; and wherever more 
amplified descriptions have been published, or reductions made 

* In the text this appears as Clcmcena, a manifest typographical error 
for Clausena as printed in the index to Burman's work. According to 
Wittstein the name is derived from P. Clauson, a Danish author of a work 
on algae published in 1632. 

» See Trimen Fl. Ceyl. 3 (1895) 358. 

'De Candolle, A., La Phytographie (1880) 401. 

332 '^he Philippine Journal of Science 1921 

from an examination of Burman's actual types, this has been 
indicated in the present paper. No botanist with a wide know- 
ledge of the Indo-Malayan flora seems to have made a critical 
examination of Burman's Flora Indica or of his herbarium with 
a view of correlating his work with that of other authors. It 
would admittedly have been best could the extant specimens have 
been examined; but it being impracticable for me to do this at 
the present time, I have attempted a preliminary interpretation 
of Burman's species, from the published data alone. Considering 
the usually short and imperfect descriptions a surprisingly high 
percentage of Burman's species can definitely be correlated with 
those of other authors without the necessity of examining his 
types ; but in some cases, indicated by an asterisk in the following 
enumeration, more definite interpretations than those here indi- 
cated cannot be made without an examination of the actual type 
in each case. 

In completing this short work I am impressed with the fact 
that many European botanists do not seem fully to realize the 
value and utility of types when interpreting insufficiently de- 
scribed species of the early authors. In many cases a few hours' 
journey, or in others merely a little correspondence would make 
available the data which would definitely fix the status of a 
species. Instead of this course, however, the unsatisfactory but 
easy method seems to have been pursued of leaving the unknown 
ones under "species incognitae," "species valde dubiae," "species 
excludendae," or other equally unsatisfactory categories. 

In order to make the present publication more generally useful 
to modem botanists I have not treated Burman's species in the 
sequence in which they were published under the Linnean system 
of classes and orders, but have rearranged them under the 
modern system of families and genera, following the Engler and 
Prantl system. 




CHAR A Linnaeus 
* CHARA sp/ 

? Ulva javanica Burm. f. Fl. Ind. (1768) 239. 

This was based on a Javan specimen with a citation of the 

^ Species marked with an asterisk in this enumeration are of more or 
less doubtful status. 

19.3 Merrill: BurmarCs Flora Indica 333 

Javanese names rompot scriboe ajer and sajor codock. The 
reference to Tremella marina Dill. Muse. 46, t. 10, f. 8, cannot 
be interpreted as the type as Burman states that his specimens 
differed from this figure in rather numerous and apparently 
essential characters. Burman's species is probably a Chara. 


USNEA Linilaeus 
♦USNEA sp. 

Lichen capillaria Burm. f. Fl. Ind. (1768) 239. "Habitat in Amboina 
& Java." 

The type is Kleinhof's Javan specimen, which perhaps repre- 
sents the same form as that figured by Rumphius as Muscus 
capillaris Rumph. Herb. Amb. 6: 86, t. J^O, f. 2. It is clearly 
a species of Usnea. 



NEPHROLEPIS RADICANS (Burm. f.) Kuhn^in Ann. Mus. Bot. Lugd.- 
Bat. 4 (1869) 285. 
Polypodium radicans Burm. f. Fl. Ind. (1768) 233, t. 66, f, S. 

A well-known species of wide distribution in the Indo-Malayan 


DAVALLIA DENTICULATA (Burm. f.) Mett. ex Kuhn Fil. Deck. (1867) 27. 
Admntum (Adianthum) denticulatum Burm. f. Fl. Ind. (1768) 236. 
The actual type is a Javan specimen collected by Pryon. 
Burman notes that this specimen differed somewhat from Plu- 
kenet Phyt. 151, t. ISO, f. i, which he cites as representing the 
same species. 


ODONTOSORIA CHINENSIS (Linn.) J. Sm. Bot. Voy. Herald (1857) 430. 
Trichomanes chinensis Linn. Sp. PI. (1753) 1096. 
Adiantum (Adianthum) chinense Burm. f. Fl. Ind. (1768) 236. 

The actual type was a Chinese specimen, and the description, 
while short, seems to apply to the very common Odontosoria 
chinensis (Linn.) J. Sm. Burman adds a reference to Plukenet 
Phyt. 10, t. h, f. 1, and to Ray Hist. 1854. 

334 The Philippine Journal of Science 1921 

BLECHNUM Linnaeus 

BLECHNUM INDICUM Burm. 1 Fl. Ind. (1768) 231. "Habitat in Java." 
This is almost certainly identical with Blechnum serrulatum 
Rich. Act. Soc. Hist. Nat. Paris 1 (1792) 114, and if this proves 
to be the case Burman's name should be adopted for this pan- 
tropic species. It is the only Malayan species of the genus to 
which Burman's description applies. 

BLECHNUM ORIENTALE Linn. Sp. PI. (1753) 1077. 

Poly podium simplex Burm. f. Fl. Ind. (1768) 235. "Habitat in Am- 

Burman apparently had a Javan specimen collected by Klein- 
hof. Lonchitis amboinica rubra Rumph. Herb. Amb. 6: 70, t. 
SO, f. 1, cited as a synonym, is clearly identical with the common 
and widely distributed Blechnum orientale Linn. 


STENOCHLAENA PALUSTRIS (Burm. f.) Bedd. Ferns Brit. Ind. Suppl. 
(1876) 234. 
Polypodium palustre Burm. f. Fl. Ind. (1768) 234. "Habitat in indiis." 

Burman apparently had a Javan specimen, as he cites the 
local Javanese name as daun peekou, although his description is a 
compiled one from pre-Linnean authors. The species is perhaps 
typified by the reference to Linn. Fl. Zeyl. 200 and to Burm. 
Thes. Zeyl. 100, t ^6. The species is a well-known one of wide 
distribution in the Indo-Malayan region. 


HEMIONITIS ARIFOLIA (Burm. f.) Moore Index (1859) 114. 

Asplenium arifolium Burm. f. Fl. Ind. (1768) 231. "Habitat in India." 

A well-known characteristic species of wide distribution in the 
Indo-Malayan region. 

ADIANTUM Linnaeus 

ADIANTUM PHILIPPENSE Linn. Sp. PI. (1753) 1094. 

Adiantum (Adianthum) lunulatum Burm. f. Fl. Ind. (1768) 235. 
^'Habitat in Malabara & Java." 

The actual type was apparently a Javan specimen, the Malabar 
reference being apparently added either from Avenka Rheede 
Hort. Malabar. 12: 72, t iO, or from Petiver Gaz. t 5U, /. 10, 
both cited as representing the species. The Linnean binomial 

19,3 Merrill: Burman's Flora Indica 335 

is the oldest one for this well-known species, which is very com- 
mon and widely distributed in the Philippines. 


CHEILANTHES TENUIFOLIA (Burm. f.) Sw. Syn. (1806) 129, 332. 
Trichomanes tenuifolia Burm. f. Fl. Ind. (1768) 237. 

The type is an actual specimen from Pryon, probably of Javan 
origin. Dryopteris cantpestris Rumph. Herb. Amb. 6: 77, t. 84, 
/. 2, is cited by Burman as representing his var. p. The 
Rumphian figure is an excellent illustration of the common and 
widely distributed Cheilanthes tenuifolia Sw. 

PTERI8 Linnaeus 

PTERIS ENSIF0RMI8 Burm. f. Fl. Ind. (1768) 230. "Habitat in Zeylona 
& Java." 

This is a common and well-known species of wide distribution 
in the Indo-Malayan region, extending to China, Polynesia, and 

PTERIS VITTATA Linn. Sp. PI. (1753) 1074. 

Polypodium trapezoides Burm. f. Fl. Ind. (1768) 233, t, 66, f. 2, 
"Habitat in Java.'' 

Burman's figure clearly represents a juvenile form of the very 
common Indo-Malayan species currently known as Pteris longi- 
folia Linn. Hieronymus considers that the Old World Pteris 
vittata Linn, is specifically distinct from the tropical American 
Pteris longifolia Linn. 


POLYPODIUM SCOLOPENDRIUM Burm. f. Fl. Ind. (1768) 232 (scolo- 
pendria), "Habitat in India." 
Polypodium phymatodes Linn. Mant. 2 (1771) 306. 

The actual type of the species is doubtless Pryon's Javan 
specimen for which the local name daun sembang is cited. Pre- 
Linnean references are given to Burnfian Thes. Zeyl. 196, M 86, 
Breynius Cent. 190, t. 98, f. 1-3, Morison Hist. 3, f. H, t. 1, /. 
17, Plukenet Mant. 82, Almath. 94, and Phyt. 404, f. 1, 5. Bur- 
man's specific name will replace the Linnean one for this very 
common and widely distributed species. Linnaeus in his de- 
scription of Polypodium phymatodes gives the same references 
as does Burman f . to Burman and to Plukenet. It is probable 
that his actual type was material transmitted to him by Burman. 

336 The Philippine Journal of Science 1921 


♦CYCLOPHORUS LONGIFOLIUS (Burm. f.) Desv. in Berl. Mag. 5 (1811) 

Acrosticum longifolium Burm. f. Fl. Ind. (1768) 228. "Habitat in 

Burman's description is the basis of Gyclophorus longifoUiis 
Desv., a species of uncertain status within the genus, but in all 
probability a synonym of C adnascens (Sw.) Desv., or C. varius 
(Kaulf.) Gaudich. The description was based on a Javan 
specimen collected by Pryon, and a reference to Aspleniunt 
longifolium Petiver Gaz. 310, t. 61, f. 2, Jf. His specific name is 
older than either of these or than Gyclophorus acrostichoides 
(Forst.) Presl, another suggested reduction of the species, but 
the exact status of Acrosticum longifolium Burm. f. cannot be 
determined from the imperfect description alone. 



GLEICHENIA LINEARIS (Burm. f.) C. B. Clarke in Trans. Linn. Soc. 
Bot. 1 (1880) 428. 

Polypodium lineare Burm, f. Fl. Ind. (1768) 235, t 66, /. 4- "Habitat 
in Java." 

Burman's actual type is a Javan specimen. He adds a refer- 
ence to Linn. Fl. Zeyl. 201. The species is a well-known and 
characteristic one of wide distribution, and Burman's excellent 
figure leaves no doubt as to the exact form intended by his 



LYGODIUM PEDATUM (Burm. f.) Sw. in Shrad. Journ. 1800' (1801) 106. 

Ophioglossvm pedatum Burm. f. Fl. Ind. (1768) 227, t. 66, f, 1, 
Ophioglossum circinnatum Burm. f. op. cit. 228. 
Lyg odium circinnatum Sw. Syn. (1806) 153. 

Ophioglossum pedatum Burm. f , was based on a Javan speci- 
men, and O. circinnatum Burm. f , on a Javan specimen with a 
reference* to Adianthunc volubili polypodioides Rumph. Herb. 
Amb. 6; 76, t. S3, and Petiver Gaz. t 6i, f. 10, Page priority 
will involve the acceptance of Burman's specific name pedatum, 
for both of his descriptions apply to a single well-known and 
widely distributed species. 

19,3 Merrill: Burman's Flora Indica 337 


♦ POLYPODIUM ADIANTHOIDES Burm. f. Fl. Ind. (1768) 234. "Habitat 

in Indiis." 

Burman's description was based on an actual specimen col- 
lected by Pryon, probably of Javan origin. The reference to 
Petiver Fil. 69 t. 2. f. 9, cannot be interpreted as the type of the 
species. The species is unrecognizable from the short descrip- 
tion alone, but it is probably no Polypodium. 

♦ POLYPODIUM GLABRUM Burm. f. Fl. Ind. (1768) 235. "Habitat in 

Java, Zeylona." 

The type is apparently a Javan specimen collected by Pryon ; 
Burman added a reference to Filix non ramosa, foliis integris 
non serratis, zeylanica minor Burm. Thes. Zeyl. (1737) 100, 
which is unrecognizable from the description and which may or 
may not represent the same species as the Javan fern. Burman's 
species is unrecognizable from the description alone. 

♦POLYPODIUM OVATUM Burm. f. Fl. Ind. (1768) 233. "Habitat in 

The very brief description "fronde petiolata alterna, laciniis 
oblongis integerrimis" was based on a Javan specimen, but the 
identity of the species cannot be determined from the description 
alone. The species cannot be interpreted from the reference to 
Petiver Fil. 38, t H, /. 2, as this characterizes the variety p of 
Burman's species. 

♦ POLYPODIUM ROSTRATUM Burm. f. Fl. Ind. (1768) 235. "Habitat in 


This species was based on an actual specimen collected by 
Pryon with a reference to Plukenet Mant. 80, t. 399, f. 4. Bur- 
man compares it to Polypodium deciossatum Linn., which is a 
Dryopteris. The species is unrecognizable from the description 
alone, but is evidently no Polypodium^. 

♦TRICHOMANES NIVEA Burm. f. Fl. Ind. (1768) 237. "Habitat in 
utraque India." 

There is little evidence tha^ Burman had a specimen except 
in his statement "in utraque India." It is probable that he 
had a specimen from India or Java. I have not seen Sloane's 
figure which Burman cites as representing the species. The 
species is one of unknown status. 

181062 6 

338 ^^6 Philippine Journal of Science 1921 



PODOCARPUS JAVANICUS (Burm. f.) comb. nov. 

Thuja ? javanica Burm. f. Fl. Ind. (1768) 202 (err. typ. 302), 

L 64, /. S. "Habitat in Java." 
Podocarpus imbricatus Blume Enum. PI. Jav. (1827) 89; Pilger in 

Engl. Pflanzenreich 18 (1903) 56. 
Podocarpus cupressina R. Br. ex Mirb. in Mem. Mus. Paris 13 (1825) 


This is clearly a form of the species commonly known as Podo-^ 
carpus imbricatus Blume, as indicated by the description of 
the upper leaves on the branchlets as lanceolate and again as 
plane and ovate-lanceolate. The figure presents a branch with 
branchlets covered with the small, imbricate, greatly reduced 
leaves, the tips of the branchlets presenting the transition stage 
to the much larger, differently shaped, distichous ones. Bur- 
man's name, being much the older, should be adopted for this 
well-known and widely distributed Malayan species. 




PANICUM Linnaeus 

PANICUM PUNCTATUM Burm. f. Fl. Ind. (1768) 26. "Habitat in India." 

This was apparently based on Gramen paniceum, multiplici 

spica, maderaspatanum Pluken. Phyt. 174, t. 191, f. 1. The 

specimen in the Sloane herbarium, British Museum, on which 

Plukenet's description was based, is extant and is the form 

described as Burman's species by Hooker f. FL Brit. Ind. 7 

(1897) 29. 


PENNISETUM GLAUCUM (Linn.) R. Br. Prodr. (1810) 195. 
Panicum glaucum Linn. Sp. PI. (1753) 56. 
Panicum americanum Linn. 1. c. 

Pennisetum typhoideum A. Rich, in Pers. Syn. 1 (1805) 72. 
Alopecurus typhoides Burm. f. Fl. Ind. (1768) 27. 

No locality is cited, the species apparently being typified by 
Gramen alopecuroides, spica maxima Indiae orientalis Pluken. 

19.3 Merrill: Burman's Flora Indica 339 

Almag. 174, t. 32, f. i. It is clearly identical with the species 
currently known as Pennisetum typhoideum A. Rich, in Pers. 
Syn. 1 (1805) 72, the generally accepted name of the species. 
However, this name is antedated by Holcus spicatus Linn. Syst. 
ed. 10 (1759) 1305, typified by the same reference to Plukenet, 
and further by Panicurrt americanum Linn. Sp. PI. (1753) 56, 
and by Panicum gUiucum Linn 1. c, the latter having page prior- 
ity. Additional synonyms are: Pennisetum spicatum R. & S., 
and Pennisetum americanum K. Schum. in Engl. Pflanzenw. 
Ostafr. B (1895) 51. Munro, Journ. Linn. Soc. Bot. 24 (1887) 
136, has shoAvn that the type of Panicum glaucum Linn, is a 
Pennisetum, not a Setaria. See Hitchcock in Am. Journ. Bot. 
2 (1915) 300, and Chase op. cit. 8 (1921) 41-49. 

SPINIFEX Linnaeus 

SPINIFEX LITTOREUS (Burm. f.) Merr. in Philip. Journ. Sci. 7 (1912) 
Bot. 229, Interpret. Herb. Amb. (1917) 92. 
Stipa littorea Burm. f. Fl. Ind. (1768) 29. ^ 

Stipa spinifex Linn. Mant. 1 (1767) 84. 
Spinifex squarrosus Linn. Mant 2 (1771) 300. 

No locality is cited, but Burman gives pre-Linnean refer- 
ences to India, Ceylon, and Amboina, and cites the Javanese 
name rompot laut; all of these refer to the species commonly 
known as Spinifex squarrosus Linn. The earliest valid specific 
name is that supplied by Burman's publication. 

ELEUSINE Gaertner 
ELEUSINE LAGOPOIDES (Burm. f.) comb. nov. 

Cynosurus lagopoides Burm. f. Fl. Ind. (1768) 29. "Habitat Coro- 

Eleusine brevifolia R. Br. in WaH. Cat. (1831) No. 3815, nomen 

nudum; Hook. f. Fl. Brit. Ind. 7 (1897) 294. 
Dactylis brevifolia Willd. Sp. PI. 1 (1797) 410. 

This is reduced in Index Kewensis to Aeluropus pubescens 
Trin., which is a synonym of Aeluropus villosus Trin. = Aelu- 
ropus lagopoides (Linn.) Trin., which Burman otherwise de- 
scribed and figured as Dactylis lagopoides; see page 340. 
Burman's Cynosurus lagopoides is apparently identical with 
the species currently known as Eleusine brevifolia R. Br., so 
that it would seem that the new combination is necessary for 
this species. Eleusine lagopoides (Burm. f.) Merr. and Aeluro- 
pus lagopoides (Linn.) Trin. are very badly confused in the 
early botanical literature. 

340 ^he Philippine Journal of Science 1921 


DACTYLOCTENIUM AEGYPTIUM (Linn.) Richt. PI. Europ. 1 (1889) 68. 
Cynosurus aegyptms Linn. Sp. PI. (1753) 72. 
Dactylis ffeniculatus Burm. f. Fl. Ind. (1768) 28, t, 12, /. 3, "Habitat 

in Java." 
Spartina geniculata Beauv. Agrost. (1812) 25. 

Although the rather poor drawing does not show the charac- 
teristic mucronate tip of the rachis, and further presents but 
three spikes, this is clearly referable to the very common species 
currently known as Dactyloctenium aegyptiacum Willd. The 
only other possibility is Eleitsine corocana (Linn.) Gaertn., 
which Burman otherwise described in the same work as Cyno- 
surus corocanus Linn. 


AELUR0PU8 LAG0P0IDE8 (Linn.) Trin. ex Thwaites Enum. PL ZeyL 

(1864) 374 (lagopodioides) . 
Dactylis lagopoides Linn. Mant. 1 (1767) 83. 
Dactylis lagopoides Burm. f. Fl. Ind. (1768) 28, t, 12, /. 2, 
Aeluropus villosics Trin. Fund. Agrost. (1820) 143; Hook. f. Fl. Brit. 

Ind. 7 (1897) 334. 

No locality is cited, but three references are given to pre- 
Linnean literature. The species was described by Linnaeus one 
year earlier as Dactylis lagopoides Linn. Burman's species may 
have included two distinct forms, but his figure is clearly the 
species currently known as Aeluropus villosu^ Trin. ; he gives a 
reference to Gramen dactyloides javanicum Garzin herb., but no 
representative of the genus Aeluropus is known from Java. 


CYPERUS Micheli 

CYPERUS R0TUNDU8 Linn. Sp. PL (1753) 45. 

Schoenus tuberostis Burm. f. Fl. Ind. (1768) 19. "Habitat in Java & 

This reduction has been made from Burman's description; 
I have not seen the illustrations in Rheede and in Sloane which 
Burman cites as illustrating his species. 

♦ CYPERUS UMBELLATUS Burm. f. Fl. Ind. (1768) 21. t 9, f, 1. 

No locality is given by Burman. Regarding this species C. 
B. Clarke in Hook. f. Fl. Brit. Ind. 6 (1893) 619 states "perhaps 
grass." The description and figure appear to be based on a 
mixture of two entirely distinct forms. The stem and leaves 

19.3 Merrill: Burman*s Flora Indica 34^ 

unmistakably represent a cyperaceous plant; the inflorescences 
are distinctly grasslike, and the statement that the peduncles 
are sheathed to the middle would indicate that the inflorescences, 
drawn as attached to the vegetative parts of a cyperaceous plant, 
are indeed the upper parts of grass culms. 


FUIRENA UNCINATA (Willd.) Kunth Enum. PL 2 (1837) 184. 
Scirpus uncinatus Willd. Sp. PI. 1 (1797) 300. 
Scirpus capitattis Burm. f. Fl. Ind. (1768) 21. "Habitat in India." 

This reduction follows C. B. Clarke in Hook. f. Fl. Brit. 
In|j[. 6 (1893) 666. Burman's specific name is much older than 
Kunth's but is invalidated in Fuirena by F. capitata Willd. 


ELEOCHARIS DULCIS (Burm. f.) Trin. ex Henschel Vita Rumph. (1833) 
186;, Merr. Interpret. Herb. Amb. (1917) 104. 
Andropogon dulds Burm. f. Fl. Ind. (1768) 219. "Habitat in India." 
Hippuris indica Lour. Fl. Cochinch. (1790) 35. 
Carex tuberosa Blanco Fl. Filip. (1837) 35. 

This was based on Cyperus dulcis Rumph. Herb. Amb. 6: 7, 
t S, f. 1. The reference to Plukenet added by Burman typifies 
the var. p and cannot be interpreted as the type of the species. 
There is no evidence that Burman had an actual specimen. 


Carex ovata Burm. f. Fl. Ind. (1768) 194 (err. typ. 294). "Java." 

This has been placed as a wholly doubtful species of the section 
Primocarex, but I suspect that it does not belong in the genus, 
but perhaps in Eleocharis or possibly Fimbristylis, for no spe- 
cies of the section Primocarex is known from Java. The ref- 
erences to figures in Plukenet and Sloane cannot be interpreted 
as the type of the species in view of Burman's statement that he 
had a Javan specimen. I have seen neither of the figures cited. 

SCLERIA Bergius 

* SCLERIA sp. 

Schoenus paniculatua Burm. f. Fl. Ind. (1768) 19. "Had. Javanls. 
ex India missa." 

Burman's description conforms better with the characters 
of Scleria hancana Miq. than with any other species known to 
me, but Rheede's illustration cited by Burman as representing 
his species certainly does not represent Miquel's species. Schoe- 
nus paniculatus Burm. f . has been referred to Scleria sumatrensis 

342 The Philippine Journal of Science 1921 

Retz. (which has strongly tessellated fruits) and to S. elata 
Thwaites (which also has tessellated fruits), while Burman 
described the fruits as "laevibus" and as "splendentia saepe 
nigricantibus." Burman's specific name is valid in Scleria and 
eventually should probably be taken up for some well-known 
Malayan species of the genus now appearing in literature under 
some other specific name. The word "ilad" is a Javanese name 
for various species of Scleria. 



RHAPHIDOPHORA LACINIATA (Burm. f.) comb. nov. c- 

Polypodium lacimatum Burm. f. Fl. Ind. (1768) 231. "Habitat in 

Java & Malabara." 
Pothos pertusus Roxb. Fl. Ind. 1 (1820) 456. 
Rhaphidophora pertusa Schott in Bonpl. 5 (1857) 45; Engl. & Krause 

in Engl. Pflanzenreich 37 (1908) 47. 

Burman states regarding his Javan specimen, for which he 
cites the local names kakajar and tally kassa, that the leaves 
lacked the perforations, but these are by no means constant in 
this species. From Burman's reference to Elettadi maravara 
Rheede Hort. Malabar. 12: 41, t 20, 21, it is clear that his species 
is a Rhaphidophora, identical with the species commonly known 
as R. pertusa (Roxb.) Schott, which extends from India to Java. 
Burman's specific name, being much the older, should be adopted. 

COM M ELI N A Plumier 
COMMELINA PAPILIONACEA Burm. f. Fl. Ind. (1768) 17 U 7, /. 1. 
"Habitat Coromandeli." 
Commelina salicifoUa Roxb. Fl. Ind. 1 (1820) 176. 

Burman's species is clearly identical with the one later de- 
scribed by Roxburgh as Commelina salicifolia and, although it 
is not mentioned by C. B. Clarke in his monograph of the Com- 
melinaceae ® his specific name should be retained for the species. 

COMMELINA NUDIFLORA Linn. Sp. PI. (1753) 41. 

Commelina diffusa Burm. f. Fl. Ind. (1768) 18, t. 7, /. 2, "Habitat 

Burman's species is clearly a form of the ubiquitous Com^ 
melina nudiflora Linn., where it was placed as a synonym by C. 
B. Clarke. 

• DC. Monog. Phan. 3 (1881) 113-324. 

19,3 Merrill: Burman's Flora Indica 343 


Commelina nervosa Burm. f. Fl. Ind. (1768) 18, t. 7, /. 5. "Habitat 

C. B. Clarke was certainly correct in placing Burman's species 
as a synonym of Commelina benghalensis Linn., a very common 
plant in the Indo-Malayan region. 


ANEILEMA MALABARICUM (Linn.) Merr. in Philip. Journ. Sci. 7 
(1912) Bot. 232. 
Tradescantia malaharica Linn. Sp. PL ed. 2 (1762) 412. 
Commelina nvdiflora Linn. Mant. 2 (1771) 177, non Linn. 1753. 
Commelina nudicaulis Burm. f. Fl. Ind. (1768) 17, 1. 18, /. 1. "Java." 
Aneilema nudifiorum R. Br. Prodr. (1810) 271. 

The figure and description clearly indicate that Burman's 
species is identical with the very common Indo-Malayan form 
currently known as Aneilemu nvdiflorum R. Br. 



MONOCHORIA VAGINALIS (Burm. f.) Presl Rel. Haenk. 1 (1827) 128. 
Pontederia vaginalis Burm. f. Fl. Ind. (1768) 80. 

Burman's species was based on pre-Linnean references to 
Rheede and to Plukenet. The same species was described un- 
der the same binomial three years later by Linnaeus Mant. 2 
(1771) 222. A common plant in the Indo-Malayan region. 



(Alpinia auct., non Linnaeus) 

LANGUAS MALACCENSIS (Burm. f.) comb. nov. 

Maranta malaccensis Burm. f. Fl. Ind. (1768) 2. "Habitat in 

Alpinia malaccensis Rose, in Trans. Linn. Soc. 8 (1307) 330, non 

Roxb., nee K. Schum. 
Alpinia nohilis Ridl. in Journ. Str. Branch Roy. As. Soc. 32 (1899) 


Burman's binomial typifies Alpinia malaccensis Rose., a species 
that has apparently been misinterpreted by modern authors; 
see Valeton in Merr. Interpret. Herb. Amb. (1917) 155. The 
type of the genus Alpinia, as described by Linnaeus, is Alpinia 
racemosa, of tropical ' America, which currently appears in 

344 ^^^ Philippine Journal of Science 1921 

botanical literature as Renealmia racemosa (L.) A. Rich. This 
is the only species of Alpinia that was known to Linnaeus; hence 
it must be the generic type. The proper application of the 
generic name Alpinia is to the numerous American species now 
known as Renealmia, the latter generic name falling as a 
Bynonym. Among the numerous synonyms of Alpinia auct., 
non Linn., Langicas is the earliest available one for the numerous 
Old World species currently but erroneously referred to Alpinia, 


GALEOLA Loureiro 
♦ GALEOLA sp. 

Cassytha comiculata Burm. f. Fl. Ind. (1768) 93, t S3, /. i. "Hab- 
itat in Java.'* 

The type was a Javan specimen collected by Kleinhof. The 
reference to Cassutha cornea Rumph. Herb. Amb. 7: 52 must 
be excluded as it refers to the mycelium of Marasmius. Dr. 
J. J. Smith, in answer to my suggestion that Burman's species 
might be a Galeola, states in lit., July 12, 1918, that he is con- 
vinced that it is either Galeola altissima Blume or G. pterosperma 
Schltr., the drawing representing the upper part of a plant. 
The "thorns" represent the roots; the **floruli imperfecti" 
probably are the young branches or young floral branches which 
often appear in the axils of the scales. 


DENDROBIUM GANINUM (Burm. f.) comb. nov. 

Epidendrum caninum Burm. f. Fl. Ind. (1768) 189. "Habitat in 

Dendrobium crumenatum Sw. in Schrad. Journ. 2 (1799) 237. 

This was apparently based on a Javan specimen, judging from 
the Javanese name cited, angrec utan, and is unquestionably the 
very common and widely distributed species currently known 
as Dendrobium crumenatum Sw. The Rumphian reference An- 
graecum caninum Rumph. Herb. Amb. 6: 105, t, Jf7, f. 2, repre- 
sents the very closely allied D. papilioniferum J. J. Sm., of the 
Moluccas. I propose to adopt Burman's name in place of 
Swartz's for this very common and well-known species. 

♦DENDROBIUM sp. § Aporum. 

Epidendrum articulatum Burm. f. Fl. Ind. (1768) 189. "In India 
ex arboribus dependet." 

19,3 Merrill: Burman's Flora Indica 345 

This was apparently based on an actual specimen collected by 
Pryon, in Java ( ?) . The only other* reference given is to Herba 
supplex quinta Rumph. Herb. Amb. 6: 111, t. 51, f. 2, which is 
Dendrobmm calceolum Roxb. Whatever Burman's species may 
ultimately prove to be, it is clearly a Dendrobium of the section 


SAL IX Tournefort 

8ALIX CHINENSIS Burm. f. Fl. Ind. (1768) 211 (err. typ. 311). "Habitat 
in China and Java." 

This would appear to be identical with the later Salix japonica 
Thunb. Fl. Jap. (1784) 24, t. 31, which is common in Japan and 
in China, and which has long been cultivated in Java. It is to 
be noted that Koorders, Exkursionsfl. Java 2 (1912) 44, cites 
among other Javani specimens in the herbaria at Leiden and 
Utrecht a specimen "Kultiviert, ohne Fundort (Herbar Javan. 
Burman fide Blume = S. Japonica).'' Burman's name, being 
the older, should be retained. 

FICUS Tournefort 
FICUS AMPELOS Burm. f. Fl. Ind. (1768) 226. "Habitat in India." 

Burman gives, in addition to his short descriptive sentence, 
first, a reference to Folium politorium Rumph. Herb. Amb. 
4: 128, t. 68, and second, one to TeregaM Rheede Hort. Ma- 
labar. 3: 79, t. 60; these two forms certainly are not conspecific. 
The species should be interpreted from the first reference and 
is, I believe, the form amply described by Koorders and Valeton, 
Bijdr. Boom. Java 11 (1906) 162, under Burman's binomial. 

♦FICUS GROSSULARIOIDES Burm. f. Fl. Ind. (1768) 227. "Habitat 

in Suratta." 

This has been reduced to Ficus heterophylla Linn. f. Suppl. 
(1781) 442, which, if the correct disposition of it, would involve 
the acceptance of Burman's specific name for the species. It 
is probable that this reduction was based on the identity of 
Valli'teregam Rheede Hort. Malabar. 3: 83, t. 62, which Bur- 
man cites as representing his variety p, and which manifestly 
represents Ficics heterophylla Linn, f . Burman's very short de- 
scription was based on an actual specimen collected by Garcin. 

346 The Philippine Journal of Science 1921 

♦ FICU8 MONTANA Burm. f. Fl. Ind. (1768) 226. '^Habitat in India.'' 

The short description was based on a Javan specimen col- 
lected by Kleinhof. The description is entirely too short and 
imperfect to permit the correct interpretation of the species. 

FICUS PADANA Burm. f. FL Ind. (1768) 226. 
Fictis toxicaria Linn. Mant. 2 (1771) 305. 

The description was based on a Sumatran specimen and is the 
form described three years later as Ficus toxicaria Linn. Lin- 
naeus cites Burman's species as a synonymL The two descrip- 
tions were doubtless based on material originating from the same 
collection. Burman's name, being the older, should be retained. 

FICUS PYRI FOLIA Burm. f. Fl. Ind. (1768) 226. "Habitat in Japonia.'' 

From the very short description it is believed that this will 

prove to be identical with Ficus erecta Thunb. Ficus (1786) 

9, in which case Burman's name should replace Thunberg's. 

FICUS SEPTICA Burm. f. Fl. Ind. (1768) 226. "Habitat in India." 

Burman apparently had a Javan specimen, as he cites the 
Javanese name siri bipar for the species, but he also adds ref- 
erences to Rumphius and to Rheede. The Rumphian reference, 
which is the first one given, represents the widely distributed 
species commonly known as Ficiis lencantatoma Poir.; but I 
believe that Burman's Ficus septica should be adopted in its 
stead, as it is the oldest valid name for this particular species. 

FICUS TSJAKELA Burm. f. Fl. Ind. (1768) 227. "Habitat in Malabara.*' 

This was based on a specimen collected by Garcin ''Ficus 
surattensis & malabarica, mori folio/* and Tsjakela Rheede Hort. 
Malabar. 3: 87, t. 6i. It has been badly confused with Ficus 
infectoria Roxb., and is really Ficus infectoria Willd., non Roxb. 
It is the valid name for a species of India and Ceylon, fully 
described and figured by King, in Ann. Bot. Gard. Calcutta 1 
(1887) 57, t. 70, who has there adjusted the synonymy, on page 
61. It is entirely different from Ficus tsiela Roxb. 



BOEHMERIA PLATYPHYLLA D. Don Prodr. Fl. Nepal. (1825) 60. 

Urtica cavdata Burm. f. Fl. Ind. (1768) 198 (err. typ. 298). "Habitat 

in Java." 
Boehmeria caudata J. J. Sm. in Koord. & Val. Bijdr. Boom. Java 12 
(1910) 706, non Sw. 

19,3 Merrill: BurmarCs Flora Indica 347 

Weddell reduced Burman's species to Boehmeria platyphylla 
D. Don. var. scabrella Wedd. in DC. Prodr. 16 ^ (1859) 311, and 
it is clearly a form of Don's species. 

BOEHMERIA NIVEA (Linn.) Gaudich. Bot. Freyc. Voy. (1826) 499. 
Urtica nivea Linn. Sp. PI. (1753) 985. 

Urtica candicans Burm. f. Fl. Ind. (1768) 197 (err. typ. 297). **Hab- 
itat in Java." 

This is cleBxly Si synonym of Boehmeria nivea (Linn.) Gaudich., 
placed by Weddel as the variety candicans Wedd. of the above 

POUZOLZIA Gaudichaud 

POUZOLZIA ZEYLANICA (Linn.) Benn. FL Jav. Rar. (1838) 67. 
Parietaria zeylanica Linn. Sp. PI. (1753) 1052. 
Parietaria indica Linn. Mant. 1 (1767) 128. 

Parietaria indica Burm. f. Fl. Ind. (1768) 211. "Habitat in Java." 
Pouzolzia indica Gaudich. Bot. Freyc. Voy. (1826) 503. 

It is suspected that both the Linnean and the Burman Pa- 
rietaria indica were based on material of similar origin. The 
species is synonymous with the earlier Parietaria zeylanica 
Linn.=Pouzolzia zeylanica (Linn.) Benn. 


DEBREGEASIA LONGIFOLIA (Burm. f.) Wedd. in DC. Prodr. 16' (1869) 
235 ". 
Urtica longifolia Burm. f. Fl. Ind. (1768) 197 (err. typ. 297). "Hab- 
itat in Java." 

This is the basis of Dehregeasia longifolia (Burm. f.) Wedd., 
a well-known species of wide distribution on the mountains of 
the Indo-Malayan region. The Philippine Dehregeasia angus- 
tifolia C. B. Rob. does not appear to be distinct from the Javan 
form, although differing slightly from Indian material. 



SYNAPHEA SPINULOSA (Burm. f.) Merr. in Proc. Linn. Soc. N. S. W. 
44' (1919) 354. 
Polypodium spinulosum Burm. f. Fl. Ind. (1768) 233, U 67, /. i. 
Synaphea polymorpha R. Br. in Trans. Linn. Soc. 10 (1810) 156. 

From the excellent figure this is manifestly the proteaceous 
Synaphea polymorpha R. Br., of western Australia, this involv- 
ing the acceptance of Burman's specific name. The figure is 
excellently matched by Australian specimens before me. It 

348 "^he Philippine Journal of Science 1921 

would seem that the locality cited by Burman as "Java'' is an 


VI SCUM Tournefort 

VISCUM ARTICULATUM Burm. f. Fl. Ind. (1768) 211 (err. typ. 311). 
"Habitat in Java." 

This species, at least in a broad sense, is of very wide distri- 
bution in the Indo-Malayan region. Aspidipcia articulata Van 
Tiegh. is a synonym. 


CELOSIA Linnaeus 

CELOSIA ARGENTEA Linn. Sp. PI. (1753) 205. 

Celosia pyramidalis Burm. f. Fl. Ind. (1768) 65, t. 25, f, 1, "Ex Java 
per semina allata * * *." 

The description and figure refer to but a single species, which 
appears to be Celosia argentea Linn, rather than C. cristata 
Linn.; in Index Kewensis the species is reduced to C. cristata 
Linn, and Allmania albida R. Br. 

AERUA Forskal 

AERUA PERSICA (Burm. f.) comb. nov. 

Iresine persica Burm. f. FL Ind. (1768) 212 (err. typ. 312), t, 60, f. 1. 
Iresine javanica Burm. f. L c. t, 60, f, 2, 

Celosia lanata Linn. Sp. PL (1753) 205, non Aerua lanata Juss. 
Aerua javanica Juss. in Ann. Mus. Paris 11 (1808) 131. 

Iresine persica Burm. f . was based on a Persian specimen and 
/. javanica Burm. f. on one from Java. Both figures and de- 
scriptions apparently are referable to a single species, the former 
species being a more robust form than the latter. 


MOLLUGO Linnaeus 

MOLLUGO LOTOIDES (Linn.) C. B. Clarke in Hook. f. Fl. Brit. Ind. 2 
(1879) 662, 776. 
Glinus lotoides Linn. Sp. PI. (1753) 463; Burm. f. Fl. Ind. (1768) 

112, t. 36, /. i. 
Mollugo hirta Thuiih. Fl. Cap. (1794) 24. 
Glinus ononides Burm. f. Fl. Ind. (1768) 113, t, 36, /. 2, 
Glinus dictamnoides Burm. f . 1. c. 

The type of Glinus ononides Burm. f . was probably a Javan 
specimen, as Burman cites Pryon as the collector and gives 

19,3 Merrill: Burman's Flora Indiea 349 

the Malay name as aud-gam pait. It is apparently a form of 
Mollugo lotoides (Linn.) C. B. Clarke, which Burman describes 
on the preceding page as Glimts lotoides Linn. The type of 
Glinits dictamnoides Burm. f . was from India and can scarcely 
be other than a form of Mollugo lotoides (Linn.) C. B. Clarke. 

MOLLUGO PENTAPHYLLA Linn. Sp. PL (1753) 89. 

Mollugo paniculata Burm. f. FL Ind. (1768) 32. ''Habitat in Zeylona/' 

Burman's species is typified by Alsine lutea ramosissima 
pentaphylla polyanthos Burm. Thes. Zeyl. (1737) 12, t. 6, f. 2. 
It is nearly the typical obovate-leaved form of Mollugo penta- 
phylla Linn. Sp. PI. (1753) 89, to which the lanceolate to nar- 
rowly lanceolate-leaved form, Mollugo stricta Linn., is usually 
reduced as a synonym. 

MOLLUGO OPPOSITI FOLIA Linn. Sp. PI. (1753) 89. 

Mollugo erecta Burm. f. FL Ind. (1768) 32. 
Mollugo triphylla Burm. f. L c. 

The reference to Burman, Thesaurus Zeylanicus (1737) 13, 
t. 7, places Mollugo erecta Burm. f . definitely as a synonym of 
Mollugo oppositifolia Linn., although it would seem that Trimen, 
FL Ceyl. 2 (1894) 271, considered it rather a synonym of the very 
different Mollugo pentaphylla Linn. I can see no reason for 
distinguishing Mollugo triphylla Burm. f . from M. erecta Burm. 
t. = M. oppositifolia Linn. Burman's material of Mollugo 
erecta was from Ceylon and that of M. triphylla was from India. 


TRIANTHEMA DECANDRA Linn. Mant. 1 (1767) 70. 

Zaleya decandra Burm. f. FL Ind. (1768) 110, t. 36, /. 1. "Coroman- 

Both binomials were doubtless based on material of similar 
origin. A well-known and widely distributed plant. 


B AS ELLA Linnaeus 

BASELLA RUBRA Linn. Sp. PL (1753) 272. 

Basella japonica Burm. f. FL Ind. (1768) 76, t. 39, /. ^. "Habitat 
in Japonia." 

This is clearly referable to the Linnean species. 

350 ^^^ Philippine Journal of Science 1921 

COMETES Linnaeus 

COMETES SURATTENSIS Linn. Mant. 1 (1767) 39. 

Cometes surattensis Burm. f. Fl. Ind. (1768) 39, t, 15, f. 5, 

This was based on Indian material. The species was de- 
scribed one year earlier by Linnaeus under the same binomial, 
the material described by both authors doubtless being parts of 
the same collection. 


NYMPHAEA Toumefort 

NYMPHAEA NOUCHALI Burm. f. Fl. Ind. (1768) 120. "Habitat Coro- 
Nymphaea pubescens Willd. Sp. PL 2 (1799) 1154. 

This is tentatively reduced in Index Kewensis to Nymphaea 
stellata Willd., but it would seem that, in spite of the descrip- 
tion of the flowers as "caeruleo," it is the same as Nymphaea 
pubescens Willd. (1797), in which case Burman's name should 
be adopted for the species. Conard, The Waterlilies, Carnegie 
Inst. Washington Publ. 4 (1905) 198, 1. 17, cites Burman's species 
as a synonym of Willdenow's on the basis of an actual specimen 
examined by him: "coll. [that is, ex herb.] Burman in India, in 
hb. Delessert." 



Menispermum glabrum Burm. f. FL Ind. (1768) 216 (err. typ. 316). 
"Habitat in Java." 

The reference to Cit-amerdu Rheede Hort. Malabar. 7: 39, t. 
21, should be excluded, as it represents Tinospora cordifolia 
(DC.) Miers, a species of wide distribution in India, but which 
does not extend to Java; Burman's species was based on an 
actual Javan specimen. I cannot determine, from the short 
description or from the Javanese name cited, daun tayonam, 
the proper status of Burman's species, except that it apparently 
belongs in the Menispermaceae ; it may prove to be the same as 
Tinospora rumphii Boerl. It is not accounted for by Diels in 
his recent monograph of the Menispermaceae, Engl. Pflanzen- 
reich 46 (1910) 1 345. 

19,3 Merrill: Burman's Flora Indica 351 


MICH ELI A Linnaeus 

MICH ELI A CHAMPACA Linn. Sp. PL (1753) 536. 

Michelia euonymoides Burm. f. Fl. Ind. (1768) 124. "Habitat in 

Burman's species is certainly only a form of the earlier 
Michelia champaca Linn. ; the Javanese name cited by Burman 
is one of those still in use in Java for the Linnean species. The 
Rumphian synonym cited represents Michelia tsiampacca Linn., 
whatever that may ultimately prove to be. I have not seen 
Sloane's figure which Burman also cites as a synonym, but it 
probably does not represent Burman's species. 



Laurus malabatrum Burm. f. FL Ind. (1768) 92. "Habitat in Mala- 
baria & Amboina." 

There is no description, the species being based on references 
to Rumphius, Rheede, and **Malabathrum rmldbaricum Garcin. 
herb." The reference to Rumphius represents Cinnamomum 
javanicum Blume. Burman's species is manifestly typified by 
the Malabar specimen and the reference to Rheede, and is hence 
identical with the form described by Solander as Laurus mala- 
bathrica Soland. in Roxb. Fl. Ind. ed. 2, 2 (1832) 297. Hooker 
f., Fl. Brit. Ind. 5 (1886) 136, was unable to place it to his 


FORTUYNIA Shuttleworth 

FORTUYNIA GARCIN I (Burm. f.) Shuttl. ex Boiss. in Ann. Sci. Nat. Bot. 
II 17 (1842) 178. 
Pelta/ria garcini Burm. f. Fl. Ind. (1768) 139, t, j^6, /. 1 (garzini). 
"Habitat in Persia." 

The species is known only from Persia. 


NASTURTIUM INDICUM (Linn.) DC. Syst. 2 (1821) 199. 

Sisymbrium indicum Linn. Mant. 1 (1767) 93. 
Sisymbrium indicum Burm. f. Fl. Ind. (1768) 140. 
Sisymbrium sinapis Burm. f. 1. c. 

352 The Philippine Journal of Science 1921 

This species is widely distributed in the Indo-Malayan region. 
Sisymbrium sinapis Burm. f . was based on Javan material, while 
that on which S, indicum Burm. f. was based was from India. 
Doubtless the material Linnaeus had in describing the species 
one year earlier was a part of the same collection. 


* ? Cardamine indica Burm. f. Fl. Ind. (1768) 140. "Habitat in Java." 

De CandoUe, who saw Burman's specimen in the Delessert 
herbarium, retained this in Cardamine, with the comment "An 
potius inter Nasturtia ad sectionem Clandestinariae referenda?" 
while Schulz, Monogr. Gattung Cardamine in Engl. Bot. Jahrb. 
32 (1903) 594, states that it is a Nasturtium. I suspect that 
it is the same as Nasturtium heterophyllum Blume. Burman's 
specific name, however, is invalidated in Nasturtium by N. indi- 
cum (Linn.) DC. 


FARSETIA CANE8CEN8 (WiUd.) comb. nov. 

HeUophila canescena Willd. Sp. PI. 3 (1800) 528. 

Arabia heliophila DC. Syst. 2 (1821) 237, Prodr. 1 (1824) 147. 

HeUophila incana Burm. f. Fl. Ind. (1768) 140, t, j^6, f, 2, non Farsetia 

incana R. Br. "Habitat in India." 
Farsetia jacquemontii Hook. f. & Th. Fl. Ind. 1 (1855) 140. 

De CandoUe's and WilldenoVs binomials are based on Bur- 
man's description; the latter's specific name being invalidated 
in Farsetia, I propose to adopt WilldenoVs specific name. Far- 
setia jacquemontii Hook, f . & Th. is clearly a synonym. 

PROSOPIS Linnaeus 

PROSOPIS SPICIGERA Linn. Mant. 1 (1767) 68. 

Prosopis spicata Burm. f. Fl. Ind. (1768) 102. "Habitat Coromandeli." 

This was described one year earlier by Linnaeus and is a well- 
known Indian species. Both descriptions were doubtless based 
on material from the same collection. 

ACACIA WiUdenow 

ACACIA TRUNCATA (Burm. f.) Hort. ex Hoffmsg. Pfl. Verz. (1824) 34. 
Adianthum truncatum Burm. f. Fl. Ind. (1768) 235, t 66, f, U* 

"Habitat in India." 
Acacia decipiens R. Br. in Ait. Hort. Kew. ed. 2, 5 (1813) 463. 

19.3 Merrill: Burman's Flora Indica 353 

Kleinhof s specimen cited by Burman is credited to Java, 
the locality being certainly an error. Captain van Alderv^ereldt 
van Rosenburgh informs me that the species does not grow in 
Java, although the reduction of Adianthum truncatum Burm. 
f. to Acacia decipiens R. Br., indicated in Index Kewensis and 
in Christensen's Index Filicum, is certainly correct. The type 
was undoubtedly from western Australia, erroneously localized 
as Javan. Polypodium spinulosum: Burm. f . = Synaphea poly- 
morpha R. Br. presents a parallel case. 

SARACA Linnaeus 

SARACA INPICA Linn. Mant. 1 (1767) 98. 

Saraca arborescens Burm. f. Fl. Ind. (1768) 85, t. 25, /. 2, "Habitat 
in Java." 

Burman's species is clearly identical with the Linnean one 
published one year earlier, both descriptions doubtless being 
based on material from the same collection. 

DIALIUM Linnaeus 

DIALIUM INDUM Linn. Mant. 1 (1767) 24. 

Dialium javanicum Burm. f. Fl. Ind. (1768) 12. "Habitat in Java." 

Burman's species is clearly identical with the one described 
one year earlier by Linnaeus; both descriptions were doubtless 
based on material of the same collection. Cortex papetarius 
Rumph. Herb. Amb. 3: 212, t. 137, which is cited by Burman as 
a synonym, must be excluded, as it represents Weinmannia 
fraxinea Sm. 

CASSIA Tournefort 

CASSIA SURATTENSIS Burm. f. Fl. Ind. (1768) 97. 
Cassia glauca Lam. Encycl. 1 (1783) 647. 

The description was based on ''Senna surattensis Garzin. herb.'' 
It is apparently correctly placed as a synonym of Cassia glauca 
Lam. in spite of Burman's description of it as ''herbaceus." 
Burman's name is much older than that proposed by Lamarck 
and should be retained. 

SOP MORA Linnaeus 
SOPHORA JAPONICA Linn. Mant. 1 (1767) 68. 

Sophora japonica Burm. f. Fl. Ind. (1768) 93. "Habitat in Japonia." 

This is a well-known species of wide distribution in China 
and Japan. Both descriptions were probably based on specimens 
from the same collection. 

181052 7 

354 The Philippine Journal of Science 1921 


CROTALARIA NANA Burm. f. Fl. Ind. (1768) 156, U j^8, f, 2. "Habitat 
in India/' 

This species is a well-known one of India and Ceylon. 

CROTALARIA PERSICA (Burm. f.) comb. nov. 

Cytisus persicus Burm. f. Fl. Ind. (1768) 163, L 51, f. 1. ''Habitat 

in Persia." 
Spartium persicum Willd. Sp. PI. 3 (1800) 931. 
Crotalaria furfuracea Boiss. Diagn. 1 ^ (1843) 6. 

De Candolle, Prodr. 2 (1825) 157, retains this as a doubtful 
species of Cytisus, giving an amplified description from Bur- 
man's specimen in the Delessert herbarium. It is apparently 
identical with Crotalaria furfuracea Boiss. 


Lupinus javanicMs Burm. f. Fl. Ind. (1768) 157. "Habitat in Java." 

The genus Lupinus is represented in Java by but three species, 
all casual and apparently recent introductions. From Burman's 
description it is quite evident that he had flowering specimens 
of Crotalaria quinquefolia Linn., in spite of the fact that the 
corolla is described as ''purpurascens." In this species of Cro- 
talaria the flowers are yellow, although the petals are often 
tinged with purple on the outside. 

CROTALARIA SERICEA Burm. f. Fl. Ind. (1768) 156. "Habitat in India." 
Crotalaria burmanni DC. Prodr. 2 (1825) 126. 

This is the basis of Crotalaria burmanni DC, the new name 
having been proposed because of Crotalaria sericea Retz. (1783) , 
which, however, manifestly does not invalidate Burman's specific 
name. Baker, in Hook. f. Fl. Brit. Ind. 2 (1876) 75, places 
Crotalaria sericea Burm. f. and C. burmanni DC. as doubtful 
synonyms of Crotalaria assamica Benth. 


ASPALATHUS ANTHYLLOIDES Linn. Sp. PI. ed. 2 (1763) 1002. 

LotiLs persicus Burm. f. Fl. Ind. (1768) 173, t. J^9, f, 3, "Habitat in 

Anthyliis asphaltoides Linn. Cent. PI. 2 (1756) 183, Amoen. Acad. 4 

(1759) 326. 
07ionis ? asphaltoides DC. Prodr. 2 (1825) 167. 

De Candolle, who examined Burman's type in the Delessert 
herbarium, referred it with doubt to the genus Ononis, with the 
comment **sed descr. et patriae dubiae remanent." The Linnean 

19.3 Merrill: Burman's Flora Indica 355 

species is South African ; it would seem, therefore, that Burman 
was in error in accrediting his species to Persia. 

ONONIS Linnaeus 

ONONIS PERSICA Burm. f. Fl. Ind. (1768) 157, «. U9, /. 1. "Habitat 
in Persia." 

This is supposed to be a synonym of Ononis sicula Guss.; 
this being so, Burman's specific name is much the older and 
should be retained. 

LOTUS Tournefort 

LOTUS GARCINI DC. Prodr. 2 (1825) 212. 

Aspalathus persica Burm. f. Fl. Ind. (1768) 155. "Habitat in Persia 
& Malabaria." 

The Persian specimen, the type of the species, is a Lotits, 
and Burman's name is the basis of Lotus garcini DC. The 
reference to Rheede Hort. Malabar. 9; 71, t. 38, must be ex- 
cluded, as it does not represent the species Burman actually 
described. The type is in the Delessert herbarium. The 
specific name persica is invalidated in Lotus by Lotus persicus 
Burm. f. 


INDIGOFERA COLUTEA (Burm. f.) comb. nov. 

Galega colutea Burm. f. Fl. Ind. (1768) 172. "Habitat in India." 
Tephrosia colutea Pers. Syn. 2 (1807) 326. 
Indigo f era viscosa Lam. Encycl. 3 (1789) 247. 

This species is currently known as Indigofera viscosa Lam., 
but Burman's specific name is older and should be retained. 

INDIGOFERA ARGENTEA Burm. f. Fl. Ind. (1768) 171. "Habitat in 

Burman's name antedates Indigofera argentea Linn. Mant. 
2 (1771) 273 and should be maintained, with Indigofera semi- 
trijuga Forsk. as a synonym. The proper name for the species 
described by Linnaeus as Indigofera argentea is /. articulata 
Gouan., for his species does not seem to have been based on 
specimens of the same collection as Burman's. 

INDIGOFERA ENNEAPHYLLA Linn. Mant. 2 (1771) 272. 

Hedysarum prostratum Linn. Mant. 1 (1767) 102, non Indigofera 

prostrata Perr. 
Hedysarum prostratum Burm. f. Fl. Ind. (1768) 168, t. 55, /. 1. 

Hedysarum prostratum of Linnaeus and of Burman were 
doubtless based on material of similar origin. It is a synonym 
of Indigofera enneaphylla Linn. 

356 3^^^ Philippine Journal of Science 1921 


TEPHR08IA VILLOSA (Linn.) Pers. Syn. 2 (1807) 329. 
Galega villosa Linn. Syst. ed. 10 (1759) 1172. 
Galega barha-jovis Burm. f. Fl. Ind. (1768) 172. "Habitat in India." 

Burman's name is a synonym of the earlier Linnean one. 

TAVERN I ERA de Candolle 
TAVERN I ERA SPARTEA (Burm. f.) DC. Prodr. 2 (1825) 339. 

Hedysarum sparteum Burm. f. Fl. Ind. (1768) 166, t. 51, f. 3 (err. 
f. 2), ^'Habitat in India." 

De Candolle examined Garcin's specimen in the Delessert her- 
barium on which Burman's species was based ; he indicates that 
this was from Persia rather than from India. Taverniera num- 
mularia DC. is probably not specifically distinct. 

DESMODIUM HETEROCARPUM (Linn.) DC. Prodr. 2 (1825) 337. 
Hedysarum heterocarpon Linn. Sp. PI. (1753) 747. 
Hedysarum siUquosum Burm. f. Fl. Ind. (1768) 169, t 55, f. 2, 

Desmodium siliquosum DC. Prodr. 2 (1825) 336. 
Desmodium polycarpum DC. op. cit. 334. 

Burman's species is clearly identical with the very common 
Indo-Malayan Desmodium heterocarpum (Linn.) DC, more com- 
monly known as D, polycarpum DC. 

DESMODIUM CAPITATUM (Burm. f.) DC. Prodr. 2 (1825) 336. 
Hedysarum capitatum Burm. f. Fl. Ind. (1768) 167, t, 5U, f, 1. 

No definite locality is given, but Burman's specimen was 
from either Java or India. This is the baisis of the well-known 
and widely distributed Desmodium capitatum (Burm. f.) DC. 

DESMODIUM TRIFLORUM (Linn.) DC. Prodr. 2 (1825) 334. 
Hedysarum triflonum Linn. Sp. PI. (1753) 749. 

Hedysarum stipulaceum Burm. f. Fl. Ind. (1768) 168, t. 5Jf, f. 2. 
"Crescit in Persia." 

Burman's species has been reduced to Desmodium triflorum 
(Linn.) DC, a very common species in niost tropical countries; 
this doubtless is the correct disposition of it, as the figure agrees 
with Desmodium triflorumi DC. 

ALYSICARPUS MONILIFER (Linn.) DC. Prodr. 2 (1825) 353. 
Hedysarum moniliferum Linn. Mant. 1 (1767) 102. 
Hedysarum monili forme Burm. f. FL Ind. (1768) 168, t. 52, /. S. 

19,3 Merrill: Burman's Flora Indica 357 

Burman's species is clearly identical with the one described 
a year earlier by Linnaeus; both descriptions were doubtless 
based on material from the same collection. It is widely dis- 
tributed in India. 

URARIA Desvaux 

URARIA CRINITA (Linn.) Desv. Journ. Bot. 1 (1813) 123. 
Hedysarum crinitum Linn. Mant. 1 (1767) 102. 

Hedysarum crinitum Burm. f. Fl. Ind. (1768) 169, t. 56. "Habitat 
in Java." 

A characteristic Indo-Malayan species, both Burman's and 
Linnaeus's binomials doubtless being based on material of simi- 
lar origin. 

PON GAM I A Ventenat 

PONGAMIA PINNATA (Linn.) Merr. Interpret. Herb. Amb. (1917) 271. 
Cytisus pinnatus Linn. Sp. PI. (1753) 741. 
Robinia mitis Linn. op. cit. ed. 2 (1763) 1044. 
Robinia javanica Burm. f. Fl. Ind. (1768) 163. "Habitat in Java.*' 

Burman's species was based on a flowering specimen, and 
in view of this fact it is scarcely possible that it is a synonym 
of Cassia sulfurea DC. = Cassia glauca Lam., to which it has 
been reduced and which Burman otherwise described as Cassia 
surattensis Burm. f. The description is entirely inadequate, 
consisting only of the statement: "caule inermi foliis pinnatis 
sex jugis, pedunculis simplicibus multifloris.'' Burman's species 
is apparently identical with the common and widely distributed 
Pongamia pinnata (Linn.) Merr., currently known as Pongamia 
glabra Vent. 

ELEIOTIS de Candolle 

ELEI0TI8 MONOPHYLLA (Burm. f.) DC. Mem. Legum. 7 (1825) 350, 
Prodr. 2 (1825) 348. 

Glycine monophyllos Burm. f. Fl. Ind. (1768) 161, t, 50, f. 2. "Hab- 
itat Coromandeli." 

Hedysarum sororium Linn. Mant. 2 (1771) 270. 

Eleiotis sororia DC. 1. c. 

Burman's specific name has priority and should be retained 
for this species. The description of Burman and that of Lin- 
naeus were doubtless based on material from the same collection. 


TERAMNUS UNGINATUS (Linn.) Sw. Prodr. (1783-87) 105. 
Dolichos uncinatus Linn. Sp. PI. ed. 2 (1763) 1019. 
Dolichos uncinatus Burm. f. Fl. Ind. (1768) 161. "Habitat in India 

358 The Philippine Journal of Science 1921 

There is no evidence that Burman's binomial was intended 
to represent Dolichos uncinatus Linn., although it probably 
represents the Linnean species which is the basis of Terammcs 
uncinatus (Linn.) Sw. Burman's actual type is clearly Klein- 
hof 's specimen, as he notes that the specimen differed somewhat 
from Plukenet's figure, Phyt. 290, t. 21Jf, f. 2, which he cites 
as representing the species. The Linnean species is typified 
by Plumier Spec. 8. ic. 221; there is no specimen in the Linnean 

PHASEOLUS Tournefort 

PHASEOLUS TRILOBATUS (Linn.) Baill. in Bull. Soc. Linn. Paris 
1 (1883) 379. 
Dolichos trilobatus Linn. Mant. 1 (1767) 101. 
Glycine tHloba Burm. f. Fl. Ind. (1768) 162. "Habitat in Java." 
Glycine triloba Linn. Mant. 2 (1771) 442. 
Dolichos trilobus Ait. Hort. Kew. 3 (1789) 30. 

In describing Glycine triloba, Linnaeus cites Burman's publi- 
cation and gives as synonyms Dolichos trilobatus Linn. Mant. 
1 (1767) 101, and Phaseolus aconitifolius Jacq. Obs. 3 (1764-71) 
2, t 52, Jacquin's species is generally considered a distinct one, 
while the proper name for the form Linnaeus described is 
Phaseolus trilobatus (Linn.) Baill. Dolichos trilobatus (Linn.) 
Baill. occurs in Java, but Jacquin's species is not reported from 
that island. 


VIGNA SINENSIS (Linn.) Savi ex Hassk. Cat. Hort. Bogor. (1844) 279. 
Dolichos sinensis Linn. Cent. PI. 2 (1756) 28. 

Dolichos catjang Burm. f. Fl. Ind. (1768) 161. "Habitat in India." 
Dolichos catjang Linn. Mant. 2 (1771) 259. 
Vigna catjang Walp. in Linnaea 13 (1839) 533. 

If Burman's species be typified by the first reference to an 
illustration, and the reference from which he took his specific 
name, namely, Malay Katjang poeti, Phaseolus minor Rumphius 
Herb. Amb. 5: 383, t 139, f. 1, it is identical with Vigna cylin- 
drica (Linn.) Skeels in U. S. Dept. Agr. Bur. PL Industry Bull. 
282 (1913) 32; Merr. Interpret. Herb. Amb. (1917) 284 VPha- 
seolus cylindricus Linn. Amoen. Acad. 4 (1759) 132]. There 
is no direct evidence that Burman had an actual specimen. The 
catjang, however, does not appear to be specifically distinct from 
the cowpea, Vigna sinensis (Linn.) Savi. 

19,3 Merrill: Bur man's Flora Indica 359 


FAGONIA Tournefort 

FAGONIA CRETICA Linn. Sp. PL (1753) 386. 

Fagonia indica Burm. f. Fl. Ind. (1768) 102, t. 34y /. i. "Habitat in 

This is apparently a synonym of the Linnean species, where 
it has long been placed as a synonym. 



MURRAYA PANICULATA (Linn.) Jack in Malay Miscel. 1' (1820) 31. 
Chalcas paniculata Linn. Mant. 1 (1767) 68. 

Chalcas camuneng Burm. f. Fl. Ind. (1768) 104. "Habitat in Java." 
Murray a exotica Linn. Mant. 2 (1771) 563. 

A variable species of wide distribution in Asia and Malaysia, 
now pantropic in cultivation. 

CLAUSENA Burman f. 

CLAUSEN A EXCAVATA Burm. f. Fl. Ind. (1768) 87 (Claucena) , 243. 
"Habitat in Java." 

This is the original publication of the genus. Burman's ori- 
ginal spelling was Claucena, apparently a typographical error 
for Clausena, as he gives it in the index to his work as Clausena. 
The name, according to Wittstein, was derived from P. Clauson, 
a Danish botanist. The species is a well-known one of wide dis- 
tribution in the Indo-Malayan region. 

TRIP HAS I A Loureiro 

TRIPHASIA TRI FOLIA (Burm. f.) P. Wils. in Torrea 9 (1909) 33. 

Limonia trifolia Burm. f. Fl. Ind. (1768) 103, t. 35 (err. typ. 34), /. 1. 

"Habitat in Java." 
Limonia trifoliata Linn. Mant. 2 (1771) 237. 
Triphasia trifoliata DC. Prodr. 1 (1824) 536. 
Triphasia aurantiola Lour. Fl. Cochinch. (1790) 153. 

A very common and widely distributed Indo-Malayan species 
for which Burman's binomial supplies the earliest specific name. 
The Linnean description was doubtless based on material of the 
same collection as that on which Burman's was based. 

360 ^'^^ Philippine Journal of Science 1921 


FERONIA LIMONIA (Linn.) Swingle in Journ. Wash. Acad. Sci. 4 (1914) 
328; Merr. Interpret. Herb. Amb. (1917) 293. 
Schinus limonia Linn. Sp. PI. (1753) 389. 
Feronia elephantum Correa in Trans. Linn. Soc. 5 (1800) 225. 
Limonia acidissima Burm. f. Fl. Ind. (1768) 102. "Habitat in India." 

Burman's binomial was probably not intended as a new "one, 
but was doubtless intended as the Linnean species of the same 
name, Sp. PL ed. 2 (1762) 554. 

*SOLANUM TRIFOLIATUM Burm. f. Fl. Ind. (1768) 57, t. 22, /. 3, 

The reduction given in Index Kewensis to Triphasia auran- 
tiola Lour, is an impossible one, the latter being described and 
figured by Burman as Limonia trifolia, Fl. Ind. (1768) 103, 
t. 35, /. 1. The two figures present nothing in common, but 
depict two entirely different plants. Solanum trifoliatum Burm. 
f. apparently represents some rutaceous plant. 


PROTIUM Burman f. 

PROTIUM JAVANICUM Burm. f. Fl. Ind. (1768) 88. "Habitat in Java." 
Amyris protium Linn. Mant. 1 (1767) 65. 

The description is based on an actual specimen, with a ref- 
erence to Tingulong Rumph. Auct. 54, t. 28, /. 1, which rep- 
resents the same species. It typifies the genus. The Linnean 
Amyris protium was doubtless based on material originating 
from the same source as Burman's type. 


SANDORICUM Cavanilles 

SANDORICUM KOETJAPE (Burm. f.) Merr. in Philip. Journ. Sci. 7 (1912) 
Bot. 237, Interpret. Herb. Amb. (1917) 308. 
Melia koetjape Burm. f. Fl. Ind. (1768) 101. "Habitat in Java." 
Sandoricum indicum Cav. Diss. 4 (1787) 359, t. 202, 203. 

Burman's species is clearly identical with the common and 
widely distributed Indo-Malayan species currently known as 
Sandoricum indicum Cav. Koetjape is the common Javanese 
name for the species. 

19,3 Merrill: Burman's Flora Indica 361 



ANTIDESMA BUNIUS (Linn.) Spreng. Syst. 1 (1825) 826. 
Stilago bunius Linn. Mant. 1 (1767) 122. 
Stilago bunius Burm. f. FL Ind. (1768) 16. ''Habitat in India." 

A very common Indo-Malayan species, described by Linnaeus 
one year earlier than by Burman and under the same binomial. 
Both descriptions were doubtless based on material of similar 

CROTON Linnaeus 

CROTON GLABRESCENS Miq. Fl. Ind. Bat. 1 ' (1858-59) 382. 

Croton castaneifolium Burm. f. Fl. Ind. (1768) 205 (err. typ. 305), 
t 64y f. ly non Linn. "Habitat in Java." 

Burman^s figure represents a true Croton and is apparently 
the same as Croton glabrescens Miq. 

CROTON OBLONGUM Burm. f. Fl. Ind. (1768) 205 (err. typ. 305), t, 6^, f. 2. 
Croton laevifolius Blume Bijdr. (1825) 603. 

No locality is cited, and the entire description consists of 
the words "foliis oblongo-ovatis integerrimis.'' Burman's mate- 
rial was probably from Java, and from the figure it is clearly 
identical with the form later described as Croton laevifoliits 
Blume. Burman's name should be retained for the species. 

CROTON TIGLIUm' Linn. Sp. PL (1753) 1004. 

? Croton racemosum Burm. f. Fl. Ind. (1768) 206 (err. typ. 306), t. 62, 

The type of this was apparently a Javan specimen, and from 
the description and figure I can suggest no other reduction 
than Croton tiglium Linn., although the description of the leaves 
as tomentose scarcely applies to the Linnean species. The form 
indicated as var. /?, based on Beenel Rheede Hort. Malabar. 
5: 7, ^. i, is certainly not congeneric with Croton racemosum 
Burm. f. 


CLAOXYLON POLOT (Burm. f.) Merr. Interpret. Herb. Amb. (1917) 200. 
Croton polot Burm. f. Fl. Ind. (1768) 205 (err. typ. 305). "Habitat in 

Amboina & Java." 
Claoxylon indicum Hassk. Cat. Hort. Bogor. (1844) 235. 

Burman based his description on an actual Javan specimen, 
which must typify his species in spite of the fact that he took 

362 The Philippine Journal of Science 1921 

his specific name from Folium urens s, polot Rumph. Herb. Amb. 
3: 217, t. IJfl. The specimen described is clearly the species 
commonly known as Claoxylon indicum Hassk. The Rumphian 
reference must be excluded, as it represeqis Laportea amplissima 
(Blume) Miq. 


CLEIDION SPICIFLORUM (Burm. f.) Merr. Interpret. Herb. Amb. (1917) 

Acalypha spidflora Burm. f. Fl. Ind. (1768) 203 (err. typ. 303), t 61, 

/. 2, "Habitat in India utraque." 
Cleidion javanicum Blume Bijdr. (1825) 613. 

The form figured and hence the type of Burman's species is 
clearly the one commonly known as Cleidion javanicum Blume, 
with staminate inflorescences. The references to Plukenet, P. 
Browne, Rheede, Burman, and Rumphius must be excluded, 
as none of these pre-Linnean citations represents the form that 
Burman f. figured and described. 

ACALYPHA Linnaeus 

ACALYPHA HISPIDA Burm. f. Fl. Ind. (1768) 203 (err. typ. 303), t. 61, 
f. 1. "Habitat in India." 

Caturus spiciflorus Linn. Mant. 1 (1767) 127, non Acalypha spiciflora 

Burm. f . 
Acalypha densiflora Blume Bijdr. (1826) 628. 
Acalypha sanderi N. E. Br. in Gard. Chron. (1896) 2: 392. 

This is a species of wide distribution in the Indo-Malayan re- 
gion in cultivation, and is also cultivated for ornamental purposes 
in Europe and America. It was later described as Acalypha 
densiflora Blume, and as recently as 1896 as A. sanderi N. E. 
Br. The earlier Caturus spiciflorus Linn. Mant 1 (1767) 127 
is invalidated by Acalypha spiciflora Burm. f. 

RICINUS Linnaeus 
RICINUS COMMUNIS Linn. Sp. PI. (1753) 1007. 

Ricinus speciosus Burm. f. Fl. Ind. (1768) 207 (err. typ. 307), t, 63, 
/. 2. "Habitat in Java." 

This is certainly a form of Ricinus communis Linn., with very 
narrow leaf-segments and unusually deep sinuses. 


RHUS Tournefort 

RHUS JAVANICA Linn. Sp. PI. (1753) 265. 

Schinus indica Burm. f. Fl. Ind. (1768) 215 (err. typ. 315). "Hab- 
itat in India." 
Rhus semialata Murr. in Comm. Getting. 6 (1784) 27, t, 3. 

i9, 3 Merrill: Bur man's Flora Indica 353 

There is nothing in the description by which the identity 
of this species can be determined with absolute certainty, it con- 
sisting only of the statement: "Foliis pinnatis, foliolis ovatis 
serratis imparl aequali, petiolis alatis subcanaliculatis." The 
type probably came from Java, which being so the species is 
clearly no Schinus, all species of which are American. It can 
scarcely be other than a synonym of Rhus javanica Linn., more 
commonly known as R. semialata Murr. 


ALLOPHYLUS TRIPHYLLUS (Burm. f.) comb. nov. 

Usubis triphylla Burm. f. Fl. Ind. (1768) 89, t 32, f. 1, "Habitat in 

Allophylus javensis Blume Rumphia 3 (1847) 183. 

Burman's species has been reduced to Allophylus cobbe (Linn.) 
Blume, sensu lat„ but the latter is limited to India and Ceylon. 
Usubis triphylla is clearly identical with Allophylus javensis 

ZIZYPHUS Tournefort 
ZIZYPHUS JUJUBA (Linn.) Lam. Encycl. 3 (1789) 318. 
Rhamnus jujuha Linn. Sp. PI. (1753) 194. 

Rhamnus nummularia Burm. f. Fl. Ind. (1768) 61, saltem pro parte. 
"Habitat in Zeylona & Java." 

It is suspected that two or more species are included in the 
references given by Burman. The Javan reference to Pryon 
is probably Zizyphus jujuba (Linn.) Lam., there being no other 
Javan species known to which the descriptive sentence applies. 
''Rhamnus persica herb. Garzin/' probably appertains to Zizy- 
phus rotundifolia Lam. The Ceylon references probably apper- 
tain to Zizyphus jujuba Lam. and Z. oenoplia Mill. 

ZIZYPHUS SPINI-CHRISTI (Linn.) VV^illd. Sp. PI. 1 (1797) 1105. 
Rhamnus spini-christi Linn. Sp. PI. (1753) 195. 

Rhamnus heterogenea Burm. f. Fl. Ind. (1768) 61. "Habitat in India 
orientali & Persia." 

Burman's species is apparently referable to Zizyphus spini- 
christi (Linn.) Willd. 

SCUTIA Commerson 
SCUTIA MYRTINA (Burm. f.) comb. nov. 

Rhamnus myrtinv^ Burm. f. FL Ind. (1768) 60. "Habitat Coroman- 

Scutia indica Brongn. in Ann. Sci. Nat. 110 (1827) 363. 
Scutia commersonii Brongn. L c. ♦ 

364 The Philippine Journal of Science 1921 

It would seem that Burman's specific name should be adopted 
for this Asiatic and Madagascar species. 


LEE A Royen 

LEEA INDICA (Burm. f.) Merr. in Philip. Journ. Sci. 14 (1919) 245. 
Staphylea ? indica Burm. f. Fl. Ind. (1768) 75, t. 2U, f. 2, "Habitat 

in Indiis." 
Aquilida sambucina Linn. Mant. 2 (1771) 211. 
Leea sambucina Willd. Sp. PL 1 (1797) 1177. 

From the description and figure Burman's species is clearly 
identical with the widely distributed form currently known as 
Leea sambucina Willd., the latter being based on the Linnean 
binomial, and the Linnean binomial in turn typified by Bur- 
man's description. 


ABUTILON Tournefort 

ABUTILON PERSICUM (Burm. f.) comb. nov. 

Sida persica Burm. f. Fl. Ind. (1768) 148, t. U7, /. i. ^'Habitat in 

Sida pohjandra Roxb. Hort. Beng. (1814) 50, Fl. Ind. ed. 2, 3 (1832) 

Abutilon polyandrum Schlecht. in Link Enum. Hort. Berol. 2 (1822) 


Burman's species is clearly identical with Abutilon poly an- 
drurrt (Roxb.) Schlecht. to which it is reduced by Masters in 
Hook. f. Fl. Brit. Ind. 1 (1874) 325. Burman's specific name 
should be adopted for the species. 

SIDA Linnaeus 

SIDA ACUTA Burm. f. Fl. Ind. (1768) 147. "Habitat in India." 

This is the earliest valid name for this very common and 
widely distributed species ; it was later described by the younger 
Linnaeus as Sida carpinifolia Linn. f. Suppl. (1781) 307. 

SIDA RACEMOSA Burm. f. Fl. Ind. (1768) 148. 
Sida mysorensis W. & A. Prodr. (1834) 59. 

No habitat is cited. This species has been reduced to Sida 
glutinosa Cav., probably from confusion of Cavanilles's species 
with Sida mysorensis W. & A. The description applies very 
closely to Sida mysorensis W. & A., and it is believed that Bur- 
man's specific name should be retained for the Indo-Malayan 
form generally referred to the latter. 

19,3 Merrill: Burman's Flora Indica 365 

HIBISCUS Linnaeus 
HIBISCUS PANDURAEFORMIS Burm. f. Fl. Ind. (1768) 151, t U7y 

No species of Althaea (Alcaea) Burman, cultivated or other- 
No locality is cited, but the type was either Indian or Javan. 

The species is a well-known one, extending from India to tropical 

Africa, Java, and tropical Australia. 

*ALCAEA INDICA Burm. f. Fl. Ind. (1768) 149. "Habitat in Java." 

No species of Althaea (Alcaea) Burman, cultivated or other- 
wise, is reported from Java, and Burman's species has not been 
considered by any modern author. It is suspected that his speci- 
men represents some species of Hibiscus, but the description 
alone is scarcely sufficient to determine this matter. 


CORCHORUS Dillenius 

CORCHORUS TRIDENS Linn. Mant. 2 (1771) 566. 

Corchorus trilocularis Burm. f. Fl. Ind. (1768) 123, t. 37, f. 2, non 
Linn. ''Habitat in India." 

Burman's description and figure apparently belong with 
Corchorus tridens Linn., rather than with C. tiHlocularis Linn. 
Mant. 1 (1767) 77, at least as the two species are interpreted 
by Masters in Hook. f. Fl. Brit. Ind. 1 (1874) 397, 398. 


ME LOCH I A Dillenius 

MELOCHIA CONCATENATA Linn. Sp. PL (1753) 675. 
Melochia corchorifolia Linn. 1. c. 

Corchorus javanicus Burm. f. Fl. Ind. (1768) 123, t. 36, f, 3. 
Melochia erecta Burm. f. op. cit. 143. 

Burman's types of both species were from Java. From the 
distinctly good figure the first is manifestly synonymous with the 
species currently known as Melochia corchorifolia Linn., which 
is a synonym of M. concatenata Linn. The drawing is erroneous 
in that all the leaves are represented as alternate. Melochia 
erecta Burm. f . is clearly synonymous with Corchorus javanicus 
Burm. f. 

♦MELOCHIA CORDATA Burm. f. Fl. Ind. (1768) 143. "Coromandeli." 

If this be a Melochia it can scarcely be other than a form of 
Melochia concatenata Linn., but the description of the flowers 
as yellow, solitary, and opposite the leaves, and the capsules as 

366 The Philippine Journal of Science 1921 

compressed scarcely conforms to the characters of the Linnean 



TETRACERA AKARA (Bnrm. f.) comb. nov. 

Calophyllum akara Burm. f. Fl. Ind. (1768) 121. "Habitat in India." 
Tetracera laevis Vahl Symb. 3 (1794) 71. 

Burman's species is typified by Akara patsjoti Rheede Hort. 
Malabar. 5: 15 ^. 8, this being an excellent illustration of the 
species commonly known as Tetracera laevis Vahl. 



CALOPHYLLUM SOULATTRI Burm. f. Fl. Ind. (1768) 121. "Habitat 
in Java." 
Calophyllum spectabile WiHd. in Ges. Naturf. Fr. Berl. Mag. 5 (1811) 

Burman's binomial is the oldest valid one for the well-known 
species described in 1811 as Calophyllum spectabile Willd.; Bur- 
man's name should be retained. 

MESUA Linnaeus 

MESUA FERREA Linn. Sp. PL (1753) 515. 

Calophyllum nagassarium Burm. f. FL Ind. (1768) 121. ''Habitat 
in Amboina & Java." 

Burman's species is a synonym of the well-known Mesuu 
ferrea Linn. Nagassarium Rumph. Herb. Amb. 7:3,^. 2, cited 
by Burman as a synonym of his species, is correctly placed. 

GARCINIA Linnaeus 

Rheedia javanica Burm. f. FL Ind. (1768) 118. "Habitat in Java." 

The description clearly applies to Garcinia, but the further 
identity of the species is scarcely determinable from the de^ 
scription alone. 



COCHLOSPERMUM GOSSYPIUM (Linn.) DC. 1 (1824) 527. 
Bomhax gossypium Linn. Syst. ed. 12 (1767) 517. 
Bombax conga Burm. f. FL Ind. (1768) 145. "Habitat in India.* 

19.3 Merrill: Burman's Flora Indica 367 

Burman's species is identical with Bombax gossypium Linn., 
which was published one year earlier. 


lONIDIUM Ventenat 

lONIDIUM SUFFRUTICOSUM (Linn.) Ging. in DC. Prodr. 1 (1824) 311. 
Viola suffruticosa Linn. Sp. PI. (1753) 937. 
Poly gala thea Burm. f. Fl. Ind. (1768) 154, excl. var. /9. 
Poly gala theezans Linn. Mant. 2 (1771) 260. 

No definite locality is given, but the species is based on two 
Ceylon references, one Javan, and perhaps one Japanese, and 
is unquestionably a mixture. If typified by the first and only 
illustrations cited, Polygala frutescens, lavandulae folio, flore 
caeruleo. Burm. zeyl. 195, t 85, it is a synonym of lonidium 
suffruticosum (Linn.) Ging., which is perhaps the best dis- 
position of it. The native name kimelala, representing the va- 
riety (3, Kimelala Javanis, quae foliis ovalibus triplo majoribus, 
is unquestionably Polygala glomerata Lour., now known in Java 
as malela; it is not the type of Burman's species. Polygala 
theezans Linn. Mant. 2 (1771) 260 is a new name for Polygala 
thea Burm. f. and should go with lonidium suffruticosum (Linn.) 
Ging. as a synonym. 



FLACOURTIA INDICA (Burm. f.) Merr. Interpret. Herb. Amb. (1917) 377. 
Gmelina indica Burm. f. Fl. Ind. (1768) 132, t. 39, /. 5. "Habitat in 

Flacourtia sepiaria Roxb. PI. Coramandel 1 (1795) 48, t. 68. 
Flacourtia ramontchi L'Herit. Stirp. Nov. (1784-85) 59, t. 30, 31. 

A common species in the Indo-Malayan region, for which 
Burman's binomial supplies the oldest valid specific name. 



PHALERIA OCTANDRA (Linn.) Baill. in Adansonia 11 (1876) 321. 
Dais octandra Linn. Mant. 1 (1767) 69. 
Dais octandra Burm. f. Fl. Ind. (1768) 104, *. 32, /. 2. "Habitat in 

Drimyspermum burmanni Decne. in Ann. Sci. Nat. II 19 (1843) 40. 

A well-known Javan species, both Burman's and Linnaeus's 
descriptions doubtless being based on material of similar origin. 

368 T^^ Philippine Journal of Science 1921 



BRUGUIERA CYLINDRICA (Linn.) Blume Enum. PI. Jav. (1828) 93; 
Merr. Interpret. Herb. Amb. (1917) 388. 
Rhizophora cylindrica Linn. Sp. PI. (1753) 443. 
Rhizophora caryophylloides Burm. f. Fl. Ind. (1768) 109. "Habitat 

in India." 
Bruguiera caryophylloides Blume Enum. PI. Jav. (1828) 93. 

A widely distributed Indo-Malayan species currently known 
as Bruguiera caryophylloides Blume. 



QUISQUALIS INDICA Linn. Sp. PI. ed. 2 (1762) 556. 

Quisqualis pubescens Burm. f. Fl. Ind. (1768) 104, t. 35, J. 2, t, 28, 
f. 2 (var. /3 glabra). "Habitat in Amboina." 

Both drawings represent the well-known Quisqualis indica 

Linn.; Burman's species can be safely placed as a synonym of 

the Linnean one. 


PSIDIUM Linnaeus 

PSIDIUM CUJAVILLUS Burm. f. Fl. Ind. (1768) 114. "Habitat in India." 
Psidium pumilum Vahl Symb. 2 (1791) 56. 
Psidium cmgustifolium Lam. Encycl. 3 (1789) 17. 

Burman's binomial is the proper name for the species more 
commonly known as Psidium pumilum Vahl. His type was 
probably from Java. Cujavillus Rumph. Herb. Amb. 1: 145, 
t, Jf9, is correctly placed as a synonym. 

EUGENIA Micheli 

EUGENIA AQUEA Burm. f. Fl. Ind. (1768) 114. "Habitat in Amboina." 

This is typified by Jambosa aquea Rumph. Herb. Amb. 1 : 126, 

t. 38, f. 2, there being no evidence that Burman had a specimen. 

The species is widely distributed, in cultivation, in the Malayan 



MELALEUCA LEUCADENDRON Linn. Mant. 1 (1767) 105 (leucadendra) . 
Myrtus leucadendra Linn, in Stickm. Herb. Amb. (1754) 9, Amoen. 

Acad. 4 (1759) 120, Syst. ed. 10 (1759) 1056, Sp. Fl. ed. 2 (1762) 

Myrtus saligna Burm. f. Fl. Ind. (1768) 116. "Habitat in Java & 


19,3 Merrill: Burman's Flora Indica 369 

This is the well-known cajuput tree, widely distributed in 
Malaysia, in cultivation. 



MEMECYLON UMBELLATUM Burm. f. Fl. Ind. (1768) 87. "Habitat in 

This is the form amply described by Trimen, Fl. Ceyl. 2 
(1894) 217, who states that Memecylon umbellatum/ Burm. f. 
was published without description. It is, however, a valid pub- 
lication, as it is typified by the excellent figure cited, Burm. 
Thesaurus Zeylanicus (1737) t 31. 

JUSSIEUA Ijinnaeus 

JUSSIEUA SUFFRUTICOSA Linn. Sp. PI. (1753) 388. 

Jussiaea tenella Burm. f. Fl. Ind. (1768) 103, t. 3J^ (err. typ. 55), 
/. 2. ''Habitat in Java." 

This can scarcely be other than a form of the Linnean species, 
in spite of Burman's description of the leaves as opposite; in 
the drawing, which is good, some are drawn as alternate, others 
as opposite. 



NOTHOPANAX SCUTELLARIUM (Burm. f.) Merr. Interpret. Herb. Amb. 
(1917) 409. 
Crassula ? Scutellaria Burm. f. FL Ind. (1768) 78. "Java." 
Aralia cochleata Lam. Encyd. 1 (1783) 224. 
Panax cochleata DC. Prodr. 4 (1830) 253. 

A species in common cultivation; its native habitat is uncer- 



RHODODENDRON LEDI FOLIUM G. Don Gen. Syst. 3 (1834) 846. 
Rhododendron burmanwi G. Don L c. 
Rhododendron rosmarinifoliuTn Dippel Handb. Laubholzk. 1 (1889) 

421, non Vidal 1886. 
Azalea rosmarinifolia Burm. f. Fl. Ind. (1768) 43, t. 3y f, 3. "Habitat 

in Japonia, colitur in Java." 

Burman's species represents the well-known Rhododendron 
ledifolium G. Don, its identity having been kindly verified for 
me by Dr. B. Hayata. Rhododendron rosmarini folium (Burm. 

181052 8 

370 ^^^ Philippine Journal of Science 1921 

f.) Dippel (1889) is invalidated by Rhododendron rosmarini- 
folium Vidal (1886), the latter being an unrelated Philippine 


EMBELIA Burman f. 
EMBELIA RIBES Burm. f. Fl. Ind. (1768) 62, t. 2S, ''Habitat in Zeylonia." 

The species is a well-known one and has been correctly inter- 
preted by subsequent authors. It is the type of the genus. 



SALVADORA PERSICA Linn. Sp. PL (1753) 122. 

Galenia asiatica Buim. f. Fl. Ind. (1768) 88,.*. Sly f. 1 (err. typ. /. 5), 
''Persia & Arabia." 

Burman's species is a synonym of the earlier Linnean one. 


JASMINUM Tournefort 

JASMINUM ANGUSTIFOLIUM (Linn.) Willd. Sp. PL 1 (1797) 36. 
Nyctanthes angustifolia Linn. Sp. PL (1753) 6. 

Nyctanthes triflora Burm. f. FL Ind. (1768) 4, t. 2. "Habitat in 

Burman's species, from his excellent figure, seems to be iden- 
tical with Jasminum angustifolium (Linn.) Willd., a species 
of India and Ceylon. Ihe mdicated place of origin, Java, is 
probably erroneous. 

JASMINUM PUBESCENS (Retz.) Willd. Sp. PL 1 (1797) 37. 
Nyctanthes pubescens Retz. Obs. 5 (1789) 9. 

Nyctanthes multiflora Burm. f. FL- Ind. (1768) 5, t. 3, /. 1. "Habitat 
in China & Malabaria." 

Burman's species is apparently referable to Jasminum pubes- 
cent Willd., a species of wide distribution in tropical Asia, 
and one frequently cultivated for ornamental purposes. Bur- 
man's specific name is apparently invalidated by Jasminum mul- 
tiflorum Roth. 


*OLEA INDICA Burm. f. FL Ind. (1768) 6. "Java." 

This was based on Javan material, but the description is 
entirely inadequate, consisting only of the statement: **foliis 
lanceolatis subtus tomentosus * * * yj^ differt ab Euro- 
paea.'* The binomial does not appear in Index Kewensis. 

19,3 Merrill: BurmarCs Flora Indica 371 



ENICOSTEMA VERTICILLATUM (Linn.) Engl. Pflanzenw. Ost-Afr. C 
(1895) 313; Engl. & Prantl Nat. Pflanzenfam. 4' (1895) 67, /. SI. 

Gentiana verticillata Linn. Syst. ed. 10 (1759) 952;, Burm. f. Fl. Ind. 
(1768) 73. "Coromandeli." 

Enicostema littorale Blume Bijdr. (1826) 848. 

Burman's species is clearly the same as the Linnean one of the 
same name, and perhaps he did not intend to describe it as a 
new one. It is commonly known as E. littorale Blume. 


CARISSA Linnaeus 

CARISSA CARANDAS Linn. Mant. 1 (1767) 52. 

Capparis carandas Burm. f. Fl. Ind. (1768) 118, 119. 
Echites spinosa Burm. f. op. cit. 69. 

Burman's Capparis carandas is clearly based on Carandas 
Rumph. Herb. Amb. 7: 57, t. 25, and is a synonym of Carissa 
carandas Linn. The same binomial appears twice in Burman's 
work, in both cases based on the same Rumphian synonym. 
Echites spinosa Burm. f. seems also to be clearly referable to 
Carissa carandas Linn. 


TABERNAEMONTANA DIVARICATA (Linn.) R. Br. ex Roem. & Schultes 
Syst. 4 (1819) 427. 

Nerium divaricatum Linn. Sp. PI. (1753) 209. 

Nerium coronarium Jacq. Coll. 1 (1786) 138. 

Nyctanthes acuminata Burm. f. Fl. Ind. (1768) 5. "In Java, Mala- 
bara, Zeylona indigena, translata a Portugalli^ ex Manilhis in 
Amboinam." ^ 

Tabemaemontana coronaria Willd. Enum. Hort. Berol. (1809) 275. 

This is clearly Tabemaemontana divaricata (Linn.) R. Br., 
a species now pantropic in cultivation and more commonly known 
as T. coronaria Willd. It occurs in Manila only as an introduced 
and rarely cultivated plant and is not a native of the Philippines. 


STROPHANTHUS CAUDATUS (Burm. f.) Kurz in Journ. As. Soc. Ben^, 
46* (1877) 257. 
Echites caudata Burm. f. FL Ind. (1768) 68, t, 26, "Habitat in Javae 

locis altioribus." 
Strophanthus dichotomus DC. in Bull. Soc. Philom. 3 (1802) 123. 

372 ^h^ Philippine Journal of Science 1921 

Burman's binomial typifies the species. It is more commonly 
known as Strophanthus dichotomies DC, and extends from 
Burma and Indo-China to Malaysia. Nerium scandens Ix)ur. Fl. 
Cochinch. (1790) 116 = Strophanthus scandens R. & S. Syst. 
4 (1819) 412 does not appear to be specifically distinct. 

VALLARIS Burman f. 
VALLARIS GLABRA (Linn.) 0. Kuntze Rev. Gen. PL (1891) 417. 
Pergularia glabra Linn. Mant. 1 (1767) 53. 

Vallaris pergulanus Burm. f. FL Ind. (1768) 51. "Habitat in Java." 
Emericia pergularia Roem. & Schultes Syst. 4 (1819) 401. 

Burman's specific name should be retained for this well-known 
species. Flos pergulanus Rumph. Herb. Amb. 6: 51, t. 29, f. 2, 
is a synonym. It is probable that both Burman's and Linnaeus's 
descriptions were based on material of similar origin. 


* APOCYNUM VINCAEFOLIUM Burm. f. FL Ind. (1768) 71. 

The references are to ''Asclepias javanica angustifolia Garzin. 
herb.'' and to an illustration of an American plant in Plukenet 
Aim. 35, t, 260, /. H, It is clear that the two are neither con- 
specific nor congeneric. The short description is based on the 
Javan plant, but there are no data by which its identity can 
be safely determined; it is clearly no Apocynum, 

* ECHITES NUMMULARIA Burm. f. FL Ind. (1768) 69, t. 28, /. 1, *'Hab- 

itat in India." 

I do not recognize the species figured ; it is, however, distinct- 
ly characteristic. The citations to pre-Linnean literature added 
by Burman probably all represent species different from the 
one figured and described by him. It probably represents some 
apocynaceous plant. 

* JASMINUM OBLONGUM Burm. f. FL Ind. (1768) 6, t, 3, f. 2, "Habitat 

in Java." 

This was described and figured from Javan material; it is 
manifestly no Jasminum, but probably represents some apocyna- 
ceous plant. 


TELOSMA Coville 
TELOSMA CORDATA (Burm. f.) comb. nov. 

Asclepias cordata Burm. f. FL Ind. (1768) 72, t. 27, f. 2, "Habitat 

in Java." 
Cynanchum odoratissimum Lour. FL Cochinch. (1790) 166. 
Pergularia odoratissima Sm. Ic. (1790-93) t, 16. 
Telosma odoratissima Coville in Contr. U. S. Nat. Herb. 9 (1905) 384. 

19,3 Merrill: Bur man's Flora Indica 373 

The data given by Burman and the distinctly good figure 
clearly indicate that his species is the same as Pergularia odoror- 
tissima Sm. It is to be noted that the Javanese name malatti 
tonquin, cited by Burman from Kleinhof, is still used in Java 
for this species, being given by Koorders as malati tungking. 
The species is a native of southeastern Asia and, as the native 
name indicates, is an introduced one in Java; it is in general 
cultivation in the Indo-Malayan region for its very fragrant 
flowers. I do not consider Pergularia minor Andr. to be other 
than a synonym of P. cordata (Burm. f.) Merr. The generic 
name Pergularia of Linnaeus is thei proper one for the African 
species long placed in Daemia, Prageluria N. E. Br. (1907) 
is a synonym of Telosma. 

TYLOPHORA INDICA (Burm. f.) comb. nov. 

Cynanchum indicum Burm. f. Fl. Ind. (1768) 70. "Narin-talie. Ja- 

vanis. D. Outgaerden e Coromandel." 
Asclejnas asthmatica Linn. f. Suppl. (1781) 171. 
Tylophora asthmatica Wight & Arn. in Wight Contrib. (1834) 51. 

This is clearly the species commonly knov^n as Tylophora 
asthmatica (Linn, f.) W. & A., as Hooker f. Fl. Brit. Ind. 4 
(1883) 45 cites ''Cynanchum indicum Herb. Burm." as a synonym 
of T. asthmatica W. & A. Burman's specific name is the oldest 
valid one for the species and should be adopted. 


* PERIPLOCA DUBIA Burm. f. Fl. Ind. (1768) 70. ^'Habitat in Java & 

This has been reduced to Cryptolepis buchanani R. & S., but 
it is evident that tv^o species are involved. The reference to 
Katu pal-valli Rheede Hort. Malabar. 9: 15, t, 11, belongs with 
the Indian Cryptolepis hucJmnani R. & S. The Javan reference 
"Paepe-sajor. Javanis. D. Kleinhof apparently typifies Bur- 
man's species, but the description is altogether too short to 
warrant a reduction of it from the description alone. 


PORANIA Burman f. 

PORANIA VOLUBILIS Burm. f. Fl. Ind. (1768) 51, t. 21, /. i. "Habitat 
in Java." 

A well-known species of wide distribution in southeastern 
Asia and Malaysia; the type of the genus. 

374 The Philippine Journal of Science 1921 


JACQUEMONTIA PANICULATA (Burm. f.) Hallier f. in Engl. Bot. Jahrb. 
18 (1894) 95. 
Ipomoea paniculata Burm. f. Fl. Ind. (1768) 50, t. 21, f. 3. ''Habitat 

in Java." 
Convolvulus parviflorus Vahl Symb. 3 (1794) 29. 

A well-known and widely distributed species, for which Bur- 
man's specific name should be retained. 

MERREMIA Dennstaedt 

MERREMIA EMARGINATA (Burm. f.) Hallier f. in Engl. Bot. Jahrb. 16 
(1892) 552. 
Evolvulus emarginatu^ Burm. f. Fl. Ind. (1768) 77, t. SO, /. 1. 

"Habitat in India." 
Convolvulus reniformis Roxb. Fl. Ind. 2 (1824) 67. 
Ipomoea reniformis Choisy Conv. Or. (1834) 64. 

A well-known and widely distributed Indo-Malayan species, 
more commonly known as Ipomoea reniformis Choisy. 

MERREMIA GEM ELLA (Burm. f.) Hallier f. in Engl. Bot. Jahrb. 16 
(1892) 552. 
Convolvulus gemellus Burm. f. Fl. Ind. (1768) 46, t.21,f,l. ^'Habitat 

in Java." 
Ipomoea gemella Roth Nov. PI. Sp. (1821) 110. 
Ipomoea chrysedeis Ker in Bot. Reg. t. 270. 

Burman's binomial is the basis of the well-known and widely 
distributed Merremia gemella Hallier f. 

MERREMIA HEDERACEA (Burm. f.) Hallier f. in Engl. Bot. Jahrb. 18 
(1894) 168. 
Evolvulus hederaceus Burm. f. Fl. Ind. (1768) 77, t. SO, f. 2. ''Habitat 

in Java." 
Ipomoea polyantha Miq. Fl. Ind. Bat. 2 (1857) 613. 

A well-known and widely distributed Malayan species. 

MERREMIA VITI FOLIA (Burm. f.) Hallier f. in Engl. Bot. Jahrb. 16 
(1892) 552. 

Convolvulus vitifolius Burm. f. Fl. Ind. (1768) 45, t. 18, /. i. 
Convolvulus angularis Burm. f. op. cit. 46, t. 19, /. 2. 

Burman's material was from Java, Merremia vitifolia (Burm. 
f.) Hallier f. being a characteristic, well-known species. I 
can see no reason for distinguishing Convolvulus angularis 
Burm. f., for Burman describes its flowers as yellow, although 
including a reference to Pryon "flore purpureo.'' The reference 
to Pryon should probably be excluded. 

19,3 Merrill: Bur man's Flora Indica 375 

IPOMOEA Linnaeus 

IPOMOEA BATATAS (Linn.) Poir. in Lam. Encycl. 6 (1804) 14. 
Convolvulus batatas Linn. Sp. PI. (1753) 154. 
Dioscorea cylindrica Burm. f. Fl. Ind. (1768) 215 (err. typ. 315). 

Burman's species is typified by Kappa-kelengu Rheede Hort. 
Malabar. 7: t. 50, which clearly represents the common sweet 
potato. Hooker f., Fl. Brit. Ind. 6 (1892) 296, is in error in 
considering that Rheede's figure represents some cucurbitaceous 
plant. There is no evidence that Burman had any botanical 
material for examination in proposing the binomial Dioscorea 


QUAMOCLIT SAGITTAEFOLIA (Burm. f.) Choisy in DC. Prodr. 9 (1845) 

Ipomoea sagittaefolia Burm. 4 FL Ind. (1768) 50, t, 18, /. 2. "Habitat 

in Java." 
Ipomoea hastata Linn. Mant. 2 (1771) 204. 
Ipomoea phaenicea Roxb. Fl. Ind. 2 (1824) 92. 
Quamoclit phaenicea Choisy Conv. Or. (1834) 51. 

This is clearly a yellow-flowered form of Quamoclit phaenicea 
(Spreng.) Choisy, which is, however, not clearly distinguishable 
from Quamoclit coccinea (Linn.) Moench. 

ARGYREIA Loureiro 

ARGYREIA MOLLIS (Burm. f.) Choisy Conv. Or. (1834) 38. 

Convolvulus mollis Burm. f. Fl. Ind. (1768) 44, L 17, '^Habitat in 

Convolvulus sericeus Linn. Mant. 1 (1767) 43, non Argyreia sericea 

Dalz. & Gibs. 

Barman's specific name is the proper one for this well-known 
Malayan species. It is suspected that both Burman's and Lin- 
naeus's descriptions were based on material of similar origin. 

ARGYREIA NERVOSA (Burm. f.) Bojet Hort. Maurit. (1837) 224. 

Convolvulus nervosus Burm. f. Fl. Ind. (1768) 48, t. 20 y f. 1. "Habitat 

Convolvulus speciosus Liun. f. SuppL (1781) 137. 
Argyreia speciosa Sweet Hort. Brit. (1827) 289. 

Burman's specific name should be retained for this well-known 
species, which commonly appears in botanical literature as Argy- 
reia speciosa Sweet. 

376 ^^^ Philippine Journal of Science 1921 



HYDROLEA ZEYLANICA (Linn.) Vahl Symb. 2 (1791) 46. 
Nama zeylanica Linn. Sp. PI. (1753) 226. 

Steria aquatica Burm. f. Fl. Ind. (1768) 73, t. 39 f. 3. "Habitat 
Coromandeli in locis aquosis.'* 

This is clearly identical with Hydrolea zeylanica (Linn.) Vahl, 
a common and widely distributed species in the Indo-Malayan 
region. The figure is poor, but the description is ample. 



TRICHODESMA ZEYLANICUM (Burm. f.) R. Br. Prodr.' (1810) 496. 

Borago zeylanica Burm. f. Fl. Ind. (1768) 41, t. U, f. 2, ''Habitat 
in Zeylona." 

A common and well-known Indo-Malayan species. It was 
described by Linnaeus three years later under the same bino- 
mial, Mant. 2 (1771) 202. 


LIPPIA Houstoun 

*LIPPIA JAVANICA (Burm. f.) Spreng. Syst. 2 (1825) 752. 

Yerhena javanica Burm. f. Fl. Ind. (1768) 12, t. 6, /. 2. "Habitat 

in Java.'* 
Zapania javamdca Lam. 111. 1 (1791) 59. 

This species is one of doubtful status. The figure is distinctly 
different from the common Lippia nodiflora (Linn.) Rich., which 
is figured on the same plate by Burman as Verbena nodiflora 
Linn. The only other species of the genus reported from Java 
is Lippia asperifolia Rich, to which Burman's description scarce- 
ly applies, although H. J. Lam in his recent treatment of the 
Verbenaceae of the Malay Archipelago states: 'Trobably this 
is the same as L. asperifolia:' Should this prove to be the case, 
Burman's specific name will replace Richard's. 

PREMNA Linnaeus 
PREMNA INTEGRIFOLIA Linn. Mant. 2 (1771) 252. 
Premna serratifolia Linn. op. cit. 253. 

Comutia corymbosa Burm. f. Fl. Ind. (1768) 132, t. Uy /. i, non 
Premma corymbosa Rottl. & Willd. ''Habitat in Zeylona." 

The two forms characterized by Linnaeus do not appear to 
be distinct. Premna integrifolia Linn, is primarily based on 

19,3 Merrill: Bur man's Flora Indica 377 

Cornutia corymbosa Burm. f . Premna corymbosa Rottl. & Willd. 
was based on Indian material with no reference to Burman's 
Cornutia corymbosa. H. J. Lam, in his recent treatment of the 
Verbenaceae of the Malay Archipelago, interprets the Linnean 
species as a polymorphous one, reducing to it, among many 
others, Premria foetida Reinw., as well as the common coastal 
form with entirely glabrous leaves. The range is given as from 
Madagascar to tropical Asia, Malaysia, and Polynesia. 

GMELINA Linnaeus 
GMELINA ASIATICA Linn. Sp. PI. (1753) 626. 

Gmelina coromandelica Burm. f. Fl. Ind. (1768) 132. "Habitat in 
India utraque." 

The Coromandel form is doubtless identical with the Linnean 
species, but the reference to Sloane probably represents an en- 
tirely different plant; I have not seen Sloane's figure. 

CLERODENDRON CALAMAT08UM Linn. Mant. 1 (1767) 90. 

Volkameria alternifolia Burm. f. Fl. Ind. (1768) 137, t. U> 

Both species were described from Javan material, Burman's 
clearly being synonymous with the one described by Linnaeus 
one year earlier. Volkameria alternifolia Burm. f. was reduced 
to Clerodendron calamatosum by Linnaeus, Mant. 2 (1771). 
Both descriptions were probably based on material of similar 

CLERODENDRON PHLOMOIDES Linn. f. Suppl. (1781) 292. 

Volkameria multiflora Burm. f. Fl. Ind. (1768) 137, t. 4.5, f. i, non 
Clerodendron multiflorum G. Don. "Habitat in Java." 

Burman's species seems to be identical with that later de- 
scribed by the younger Linnaeus, but his specific name is invalid 
in Clerodendron, 


SO L A N U M Linnaeus 
SOLAN UM SURATTENSE Burm. f. Fl. Ind. (1768) 57. 

Solanum xanthocarpum Schrad. & We'ndl. Sert. Hanov. 1 : S, t. 2; C. B. 
Clarke in Hook. f. Fl. Brit. Ind. 4 (1883) 236, cum syn. 

The specimens on which Burman's description was based were 
from Surat, India, collected by Garcin. The species is unac- 
counted for in Clarke's treatment of the Solanaceae of British 
India, but the description clearly applies to Solanum xantho- 
carpum Schrad. & Wendl., which Burman's binomial displaces. 

378 ^'^^ Philippine Journal of Science 1921 

The pre-Linnean references to Dillenius, Plukenet, and Ray 
probably do not belong here ; and from one of those Dunal, who 
placed Burman's species as a synonym of Solanum virginianum 
Linn., doubtless interpreted Barman's species. 


BALLOTA Linnaeus 

BALLOTA PERSICA (Burm. f.) Benth. Lab. Gen. Sp. (1836) 598, DC. 
Prodr. 12 (1848) 520. 

Molucella persica Burm. f. Fl. Ind. (1768) 128, t. 38, f. 2, 

Bentham's amplified description is based on Barman's specimen 
in the Delessert herbarium. 


ANISOMELES INDICA (Linn.) 0. Kuntze Rev. Gen. PI. (1891) 512. 
Nepeta indica Linn. Sp. PL (1753) 571. 

Anisomeles ovata R. Br. in Ait. Hort. Kew. ed. 2, 2 (1811) 364. 
Ballota disticha Linn. Mant. 1 (1767) 83. 
Ballota disticha Burm. f. Fl. Ind. (1768) 126. 
Marruhium indicum Burm. f. op. cit. 127. 

Ballota disticha Barm. f. was from Java and Coromandel and 
was described one year earlier by Linnaeus under the same bi- 
nomial; both descriptions were doubtless based on material of 
similar origin. Marrubium indicum Burm. f. was from Ceylon 
and Java ; and from the excellent figure cited, Burm. Thes Zeyl. 
(1737) 153, t, 71, f. 1, it is, like Ballota disticha, clearly identical 
with the very common and widely distributed Anisomelis indica 
(Linn.) 0. Kuntze. 

POGOSTEMON Desfontaines 

POGOSTEMON BENGHALENSE (Burm. f.) 0. Kuntze Rev, Gen. PL (1891) 

Origanum henghalense Burm. f. Fl. Ind. (1768) 128, t. S8, f. 3. 

"Habitat in Benghalia." 
Pogostemon plectranthoides Desf. in Ann. Mus. Paris 2 (1803) 155. 

Bentham, in DC. Prodr. 12 (1848) 151, cites Burman's species 
with certainty as a synonym of Desfontaines's Pogostemon plec- 
tranthoides, adding that the figure is very poor. Burman's 
specific name should be retained. 


DYSOPHYLLA AURICULARIA (Linn.) Blume Bijdr. (1826) 826. 
Mentha auricularia Linn. Mant. 1 (1767) 81. 
Mentha foetida Burm. f. Fl. Ind. (1768) 126. "Habitat in Java." 

19.3 Merrill: Burman's Flora Indica 379 

The form Burman described was characterized one year 
earlier by Linnaeus as Mentha auricularia. Majqina foetida 
Rumph. Herb. Amb. 6: 41, t. 16, /. 2, is correctly placed as a 
synonym by both Linnaeus and Burman. Burman's species was 
reduced to Mentha foetida by Linnaeus Mant. 2 (1771). 


ACROCEPHALUS INDICUS (Burm. f.) 0. Kuntze Rev. Gen. PI. (1891) 511. 
Ptiinella indica Burm. f. Fl. Ind. (1768) 130. 
Ocimum capitatum Linn. f. Suppl. (1781) 276. 
Acrocephalus capitatics Benth. in Wall. PI. As. Rar. 2 (1831) 18. 

Burman's species was based on a Javan specimen, ''Brunella 
javanica D. Kleinhof.'' It is commonly known as Acrocephalus 
capitatus Benth., but Burman's specific name should be retained. 


Ocimum tenuifiornm Linn. Sp.* PI. (1753)' 597. 
Geniosporum prostratum Benth. in Bot. Reg. sub. t 1300, et in Wall. 

PI. As. Rar. 2 (1831) 18. 
Thymus indicus Burm. f. Fl. Ind. (1768) 129. "Habitat Coromandeli." 

This species is currently known as Geniosporum prostratum 
Benth., but it would seem that the Linnean specific name should 
be adopted for it. 

OC I M U M Linnaeus 

OCIMUM SANCTUM Linn. Mant. 1 (1767) 85. 

Ocymum inodorum Burm. f. Fl. Ind. (1768) 130. "Habitat in India." 

Burman's species is apparently a synonym of Ocimum sanctum 
Linn., where it is definitely placed by Bentham. The figure in 
Burman Thesaurus Zeylanicus (1737) 175, t. SO, f. 2, is good for 
the Linnean species, and from this citation the younger Burman 
took his specific name; the Javan Sulassi puti-utan, judging 
from this native name, might be Ocimum basilicum Linn. 


♦SCUTELLARIA ? JAPONICA Burm. f. Fl. Ind. (1768) 130. ^'Habitat 
in Japonia, Java." 

Two species are involved, but the form actually described 
by Burman was the Javan plant. It has been reduced by Ben- 
tham in part to Plectranthus coetsa D. Don, which is not known 
from either Java or Japan; and in part to Melissa parviflora 
Benth., likewise definitely known from neither Japan nor Java, 
although Bentham actually examined the species in Burman's 

380 ^'^^ Philippine Journal of Science 1921 

herbarium. A critical examination of the type material is de- 



SCHWEINFURTHIA PAPILIONACEA (Linn.) Boiss. Fl. Orient. 4: 387. 
Antirrhinum papilionaceum Linn. Mant. 1 (1767) 86. 
Antirrhinum papilionaceum Burm. f. Fl. Ind. (1768) 131, t. 39, f. 2. 

"Habitat in Persia." 
Schweinfurthia sphaerocarpa A. Braun in Sitzb. Ges. Naturf. Fr. 20 
(1866) 24. 

Both Burman's and Linnaeus's binomials were probably based 
on material from the same collection, the latter's publication 
having priority by one year. 


ARTANEMA LONGI FOLIA (Linn.) comb. nov. 

Columnea longifolia Linn. Mant. 1 (1767) 90. 

Sesamum javanicum Burm. f. Fl. Ind. (1768) 133. "Habitat in India.'* 
Archimenes sesamoides Vahl Symb. 2 (1791) 71. 
Artanema sesamoides Benth. Scroph. Ind. (1835) 39. 
Artanema longiflora Wettst. in Engl. & Prantl Nat. Pflanzenfam. 
4"'' (1891) 79. 

The Linnean name is one year earlier than Burman's, and 
both are much earlier than that proposed by Vahl on which 
Bentham's binomial is based. It is suspected that both Columnea 
longifolia and Sesamum javanicum were based on material of 
similar origin. 


(Vandellia Linnaeus) 

LINDERNIA CRUSTACEA (Linn.) F. Muell. Census (1882) 97. 
Capraria Crustacea Linn. Mant. 1 (1767) 87. 
Capraria Crustacea Burm. f. Fl. Ind. (1768) 133. 
Capraria uniflora Burm. f. 1. c. i. IJ^ (err. typ. 19), f. 3. 
Vandellia Crustacea Benth. Scroph. Ind. (1835) 35. 

Capraria Crustacea Burm. f. is identical with the species de- 
scribed one year earlier by Linnaeus under the same binomial; 
both descriptions were doubtless based on material of similar 
origin. For Capraria uniflora Burman cites no locality; but, 
judging from the figure, it is synonymous with Capraria Crus- 
tacea lAnn. --= Lindernia Crustacea (Linn.) F. Muell. 

15, a Merrill: Burman's Flora Indica 3gl 

ILYSANTHES Rafinesque 
(Bonnaya Reichenbach) 

ILYSANTHES ANTIPODA (Linn.) Merr. Interpret. Herb. Amb. (1917) 467. 
Ruellia antipoda Linn. Sp. PL (1753) 635. 
Ruellia anagallis Burm. f. Fl. Ind. (1768) 135. "Habitat in Java and 

Gratiola veronicae folia Retz. Obs. 4 (1786) 8. 
Bonnaya veronicaefolia Spreng. Syst. 1 (1825) 41. 

Burman's species is clearly identical with the very common 
and widely distributed form currently known as Bonnaya vero- 
nicaefolia Spreng. Crusta ollae majj»r Rumph. Herb. Amb. 5: 
460, t. 170, /. 2, cited by Burman as representing his species, is 
correctly placed as a synonym. 



*CRESCENTIA OVATA Burm. f. Fl. Ind. (1768) 132. "Habitat in Java." 
Crescentia cucurbitina Linn. Mant. 2 (1771) 250. 

If the current reduction of Burman's species be correct it 
will involve the acceptance of his specific name. However, the 
only species of the genus reported from Java is Crescentia cujete 
Linn., and Burman's description is so short and imperfect 
that it is impossible to determine from it alone to which of the 
Linnean species it applies. All the species of the genus are 
natives of tropical America. 



DYSCHORISTE MADURENSIS (Burm. f.) O. Kuntze Rev. Gen. PL (1891) 

Justicia madurensis Burm. f. FL Ind. (1768) 9, t. -4, /. S, "Habitat 

in Madura." 
Ruellia littoralis Linn. f. Suppl. (1781) 289. 
Calophanes littoralis T. Anders, in Thw. Enum. PL Zeyl. (1859-64) 


This is the basis of Dyschoriste madwrensis (Burm. f.) 0. 
Kuntze, of which Calophanes littoralis T. Anders. {Ruellia lit- 
toralis Linn, f.) is a synonym. 

DYSCHORISTE ERECTA (Burm. f.) O. Kuntze Rev. Gen. PL (1891) 485. 
Ruellia erecta Burm. f. FL Ind. (1768) 135, t. Al, /. 3, "Habitat 

in India." 
Calophanes nagchana Nees in DC. Prodr. 11 (1847) 109. 

382 "^^^ Philippine Journal of Science 1921 

This is the basis of Dyschoriste erecta (Burm. f.) 0. Kuntze, 
of which Calophanes nagchana Nees is apparently a synonym. 


HEMIGRAPHIS ALTERNATA (Burm. f.) T. Anders, in Journ. Linn. Soc. 
Bot. 7 (1864) 114. 
Ruellia alternata Burm. f. Fl. Ind. .(1768) 135. ''Habitat in Java." 
Ruellia blumeana Nees in DC. Prodr. 11 (1847) 149. 
Ruellia repanda Blume Bijdr. (1825) 794, non Linn. 
Hemigraphis blumeana Boerl. Handl. Fl. Nederl. Ind. 2 ' (1899) 658. 

The reference to Prunella silvestris alba Rumph. Herb. Amb. 
6: 31, t. 13 y f. 2, should be excluded, as it represents Hemigra- 
phis reptans K. Schum. var. glaucescens Hallier f. The refer- 
ence to Rheede Hort. Malabar. 9: 115, t, 59, should also be 

RUELLIA Plumier 
RUELLIA REPENS Linn. Mant. 1 (1767) 89. 

Ruellia repens Burm. f. Fl. Ind. (1768) 135, t. J,l, f. 2 (err. typ. /. i). 
"Habitat in Java." 

I consider that Burman's figure represents Ruellia repens as 
described one year earlier by Linnaeus, although C. B. Clarke, 
Journ. As. Soc. Beng. 74^ (1907) 649, considered that it rep- 
resents a plant not of the genus Ruellia, The probability is 
very great that both descriptions were based on material of 
similar origin. 

BLEPHARIS PERSICA (Burm. f.) O. Kuntze Rev. Gen. PI. (1891) 483. 
Ruellia persica Burm. f. Fl. Ind. (1768) 135. "Habitat in Persia." 
Acanthus edulis Forsk. Fl. Aeg.-Arab. (1775) 115. 
Blepharis edulis Pers. Syn. 2 (1807) 180. 

Burman's specific name, being much older than ForskaFs, 
should be retained for this species. 

(1821) 320. 
Acanthus maderaspatensis Linn. Sp. PL (1753) 639. 
Acanthus ciliaris Burm. f. FL Ind. (1768) 139, t. 1^2^ f. 3. ^'Habitat 

in Zeylona." 
Blepharis bo erhaaviae folia Pers. Syn. 2 (1807) 180. 

Burman's species is clearly identical with the Linnean one. 


ANDROGRAPHIS PANICULATA (Burm. f.) Nees in Wall. 1*1. As. Rar. 3 
(1832) 116. 
Justicia paniculata Burm. f. Fl. Ind. (1768) 9. ^'Habitat in Malabaria 
& Zeylonia." 

19.3 Merrill: Burman's Flora Indica 383 

This is the basis of the well-known Andrographis paniculata 
(Burm. f. ) Nees. 

PERISTROPHE HYSSOPI FOLIA (Burm. f.) comb. nov. 

Dianthera hyssopifolia Burm. f. Fl. Ind. (1768) 11, t, 5, /. 2, 

"Habitat in Java." 
Justicia salicifolia Blume ex Steud. Nomencl. ed. 2 1 (1841) 839. 
Peristrophe salicifolia Hassk. Cat. Hort. Bogor. (1844) 152. 

The description and excellent figure clearly indicate that Bur- 
man's species is the same as the one later described by Blume. 
Peristrophe acuminata Nees, P. blumeana Nees, and Justicia 
roxburghiana Blume are probably synonyms of the same species. 


CLIN ACANTHUS NUTANS (Burm. f.) Lindau in EngL Bot. Jahrb. 18 
(1894) 63. 

Justicia nutans Burm. f. Fl. Ind. (1768) 10, t. 5, f, 1. "Habitat 

in Java." 
Clinacanthus burmanni Nees in DC. Prodr. 11 (1847) 511. 
Justicia fulgida Blume Bijdr. (1825) 784. 

A well-known species extending from Siam to Hainan, the 
Malay Peninsula, Borneo, and Java. 

JUSTICIA Houstoun 

JUSTICIA GENDARUSSA Burm. f. Fl. Ind. (1768) 10; Linn. f. Suppl. 
(1781) 85. "Crescit in Malabara, Amboina & Java." 

A common and well-known species of wide distribution in the 
Indo-Malayan region. 

* JUSTICIA MORETIANA Burm. f. Fl. Ind. (1768) 10. "Crescit in variis 
locis Indiae arenosis." 

This is typified by Adhatoda flore solitario, etc., Burm. Thes. 
Zeyl. (1737) t 3, f. 1, and is clearly a Justicia, Trimen, Fl. 
Ceyl. 3 (1895) 335, states that the figure, although good, does 
not agree with any known Ceylon species. The synonym ''More- 
tiana Rumph. 6, p. 53, t. 23,'' must be excluded, as it represents 
Ruellia repens Linn. 


OLDEN LAN Dl A Linnaeus 

Oldenlandia tenuifolia Burm. f. FL Ind. (1768) 37, t. i^, /. 1, "Hab- 
itat in Java." 

This appears to be a form of Oldenlandia corymbosa Linn.', 
with the inflorescences reduced to a single long-pedicelled flower. 

384 ^^^ Philippine Journal of Science 1921 

or it is possibly a reduced form of Oldenlandia herbacea (Linn.) 
Roxb. (0. heynii G. Don). 

DENTELLA REPENS (Linn.) Forst. Char. Gen. (1776) 26, t. 13. 
Oldenlandia repens Mant. 1 (1767) 40. 

Oldenlandia repens Burm. f. Fl. Ind. (1768) 38, t. 15, /. 2. "Amat 
loca paludosa Coromandeli." 

It is not certain that Burman's species is identical with the 
Linnean one, although there is nothing in the description or 
in the rather indefinite figure contrary to this interpretation 
of it. 


BORRERIA OCYMOIDES (Burm. f.) DC. Prodr. 4 (1830) 544. 

Spermacoce ocymoides Burm. f. Fl. Ind. (1768) 34, t IS, /. 1. '*Sajor 
Babi Javanis." [That is, the type from Java, with the Javanese 
name sajor babi.'] 

The range of this species outside of Java is more or less con- 
fused by species of other authors that have apparently been 
erroneously reduced to it. It is, however, apparently of wide 
distribution in the Malayan region. 

PAEDERIA Linnaeus 
PAEDERIA FOETIDA Linn. Mant. 1 (1767) 52. 

Apocynum foetidum Burm. f. Fl. Ind. (1768) 71. "Habitat in India.'' 

While Paederia tomentosa Blume is generally recognized as 
a distinct species, there seenx to be no constant characters by 
which it can be distinguished from the Linnean one. 


BLASTANIA Kotschy and Peyr. 
BLASTANIA GARCINI (Burm. f.) Cogn. in DC. Monog. Phan. 3 (1881) 629. 
Sicyos garcini Burm. f. Fl. Ind. (1768) 211 (err. typ. 311). ^'Habitat 

in Zeylona.*' 
Sicyos garcini Linn. Mant. 2 (1771) 297. 
Ctenolopis garcini C. B. Clarke in Hook. f. Fl. Brit. Ind. 2 (1879) 629. 

A well-known species of southern India and Ceylon. 
LOBELIA Plumier 
* LOBELIA PUMILA Burm. f. Fl. Ind. (1768) 186, t. 60, f, 3. 

I suspect that this is a true Lobelia in spite of the leaves being 
described and figured as opposite; it is perhaps near Lobelia 
zeylanica Linn. The type was from Coromandel. 

19,3 Merrill: Burman's Flora Indica 385 


VERNON I A Schreber 

VERNONIA CINEREA (Linn.) Less, in Linnaea 4 (1829) 291. 
Conyza cinerea Linn. Sp. PI. (1753) 862. 

Conyza ivaefoUa Burm. f. Fl. Ind. (1768) 180, t, 58, f. ^. "Habitat 
in Java." 

While the figure is not particularly good for Vernonia cinerea 
Less., I cannot suggest any other reduction of Burman's species. 
The description consists merely of the statement : **f oliis alternis 
lanceolatis repando-dentatis.'' 

EUPATORIUM Tournefort 

Erigeron pisonis Burm. f. Fl. Ind. (1768) 180, t. 59, f. 2. "Brasilia." 

This was based on a Brazilian specimen collected by Piso, 
and is apparently a Eupatorium, Dr. B. L. Robinson, of the 
Gray Herbarium, who has an extensive knowledge of the trop- 
ical American Compositae, informs me that he does not rec- 
ognize the species, but that the habit, foliage, and to some extent 
the inflorescence recall Eupatorium hallotae folium HBK. and 
Conyza anomala DC, although Burman's positive statement 
*'Caul. * * * glaberrimus" does not fit either of these spe- 
cies, since both always have a perceptible pubescence on the 

Ml KAN I A Willdenow 

Ml KAN I A SCAN DENS (Linn.) Willd. Sp. PI. 3 (1804) 1743. 
Eupatorium scandens Linn. Sp. PI. (1753) 836. 

Eupatorium cordatum Burm. f. Fl. Ind. (1768) 176, t. 58, f. 2. 
"Habitat in Java & Vera Cruce." 

Burman's species is clearly referable to the common pantropic 
Mikania scandens (Linn.) Willd. 

BLUMEA de Candolle 

BLUMEA LACERA (Burm. f.) DC. Prodr. 5 (1836) 436. 
Conyza lacera Burm. f. Fl. Ind. (1768) 180, t. 59, f, 1. 

This was based on a Javan specimen and is supposedly the 
basis of Blumea lacera (Burm. f.) DC, Burman's species being 
indicated by de Candolle as var. burmanni DC. 

EPALTES Cassini 

EPALTES DIVARICATA (Linn.) Cass, in BulL Soc. Philom. (1818) 139. 
Ethulia divaricata Linn. Mant. 1 (1767) 110. 
Ethulia divaricata Burm. f. Fl. Ind. (1768) 176. "Coromandeli." 

181052 9 

386 I'he Philippine Journal of Science 1921 

Both descriptions were doubtless based on material of similar 

SPHAERANTHUS AMARANTHOIDES Burm. f. Fl. Ind. (1768) 186. 

The type was a specimen from Coromandel. The species 
is a well-known one of India and Ceylon. 

ANVILLEA de Candolle 
ANVILLEA GARCINI (Burm. f.) DC. Prodr. 5 (1836) 487. 

Anthemis gardni Burm. f. Fl. Ind. (1768) 183, t. 60, f. 1 (as Buthal- 
mum garcini Burm. f.). 

This is the basis of Anvillea garcini (Burm. f.) DC. and is 
the type of the genus. It was based on a specimen from Persia, 
collected by Garcin. 

GYNURA Cassini 
GYNURA Bl FLORA (Burm. f.) comb. nov. 

Senecio biflorus Burm. f. Fl. Ind. (1768) 181. ''Habitat in Java." 

From the Javanese name cited, blontas China, it is suspected 
that this was a cultivated plant in Java, of Chinese origin. The 
description is definite and applies unmistakably to the species 
currently known as Gynura pseudo-china DC. 

SENECIO Tournefort 

SENECIO TENUIFOLIUS Burm. f. Fl. Ind. (1768) 181, L 60, f. 4. "Habitat 
in Java." 

This is a sufficiently well-known species of India and Java, of 
which Senecio multifidus Willd. is a synonym. 

♦SENECIO CORONOPIFOLIUS Burm. f. Fl. Ind. (1768) 181, t 60, /. 5. 
"Habitat in Java." 

Senecio javanicus Willd. Sp. PL 3 (1800) 1984. 

Senecio javanicus Willd. is merely a new name for Burman's 
species. It is a species of doubtful status and is probably no 


LIGULARIA TUSSILAGINEA (Burm. f.) Makino in Bot. Mag. Tokyo 18 

(1904) 52. 
Arnica tussilaginea Burm. f. Fl. Ind. (1768) 182. 
Tussilago japonica Linn. Mant. 1 (1767) 113, non Ligularia japonica 

(Thunb.) Less. 
Senecio kaempferi DC. Prodr. 6 (1837) 363. 
Ligularia kaempferi Sieb. & Zucc. Fl. Jap. 1 (1835) 77, t. 35. 
Senecio tussilaginea O. Kuntze Rev. Gen. PI. (1891) 364. 
Farfugium kaempferi Benth. Fl. Hongk. (1861) 191. 

19' 3 Merrill: BurmarCs Flora Indica 387 

Burman's species was based on Japanese material, possibly 
on the same collection that yielded the type of Tussilago japonica 
Linn. The latter typifies Senecio kaempferi DC. Burman's 
specific name should be retained for the species. 

EMILIA Cassini 

EMILIA JAVANICA (Burm. f.) C. B. Rob. in Philip. Journ. Sci. 3 (1908) 
Bot. 217. 

Hiefacium javanicum Burm. f. Fl. Ind. (1768) 174, t. 57 y f. 1. 

Prenanthes javanica Willd. Sp. PI. 3 (1800) 1534. 
Sonchics javanicus Spreng. Syst. 3 (1826) 648. 
Emilia flammea Cass, in Diet. Sci. Nat. 14 (1819) 406, t. 5. 

Burman's specific name should be retained for this species, 
which is allied to, but distinct from, Emilia sonchi folia DC. 

SONCHUS Linnaeus 

* SONCHUS sp. 

? Senecio auriculatus Burm. f. Fl. Ind. (1768) 181. "Habitat in Java." 

This is perhaps a species of Sonchus, possibly S. oleraceus 
Linn, or S. arvensis Linn.; but the description is too short to 
warrant a more definite reduction from it alone. 


* LAPSANA JAPONICA Burm. f. Fl. Ind. (1768) 174. "Habitat in Japo- 


This is clearly no representative of the genus Lapsana. The 
type was from Japan, and Dr. B. Hayata informs me that 
neither he nor any of the other Japanese botanists at Tokyo 
recognizes the species from the description. 

* ERIGERON DENTICULATUM Burm. f. Fl. Ind. (1768) 180. 

The type of this genus was a Javan specimen. From the very 
short description it is suspected that it is a synonym of Pluchea 
indica (Linn.) Less., but the correctness of this suggested re- 
duction can scarcely be determined from the description alone. 

* ERIGERON INDICUM Burm. f. Fl. Ind. (1768) 180. "Habitat in Java." 

This is probably not an Erigeron, but Burman's description 
is altogether too short to warrant a definite reduction of it 
from the description alone. 

*ACER PLATANUS Burm. f. Fl. Ind. (1768) 221. "Habitat in India." 

Pax, in Engl. Pflanzenreich 8 (1902) 76, states regarding 
this species **Vix Aceris species.'' There is nothing in the de- 

388 '^^^ Philippine Journal of Science ' ^ 

scription by which its identity can be determined, as Burman % 

briefly describes only the leaves. 

*ACER JAVANICUS Burm. f. Fl. Ind. (1768) 221. "Habitat in Java." 

Pax makes the same comment for this species as for the pre- 
ceding one. There is nothing in the description by which its 
identity can be even surmised. The leaves only are described. 

* LUDWIGIA (LUDWIGHIA) TRIFOLIA Burm. f. Fl. Ind. (1768) 37. 

"Doedoek Javanis." [That is, type from Java, with the Javanese 
name doedoekJ] 

The Javanese name doedoek properly belongs with Lumnitzera 
littorea (Jack) Voigt, but Burman's description scarcely applies 
to this. I do not recognize the species or the group to which it 
belongs from the short and imperfect description. 

* MYRICA ASPLENIFOLIA Burm. f. Fl. Ind. (1768) 211 (err. typ. 311). 

"Habitat in Indiis." 

This was doubtless intended to represent the Linnean species 
of the same name, published in 1753, as Burman gives the same 
reference to Plukenet as does Linnaeus. The Linnean species 
is Comptonia peregrina (Linn.) Coulter, of North America. It 
would seem that at least the locality cited is an error on the part 
of Burman. 

♦RHAMNUS VITIS IDAEA Buxm. f. Fl. Ind. (1768) 61. "Habitat in 
Zeylonfc & Java." 

The reduction of this to Breynia rhamnoides Muell.-Arg. is 
not entirely satisfactory, as the descriptive sentence and most 
of the pre-Linnean references call for spiny branches. The 
figure cited by Burman f., Burman Thes. Zeyl. (1737) 198, t 
88, is, however, an excellent representation of Breynia rharn^ 
noides (Willd.) Muell.-Arg. It is suspected that the several 
references refer to several different species. The actual type 
is apparently the Javan specimen for which the local name 
boa massi is given. 

■ A f 


Vol. 19, No. 4 

October, 1921 

The Philippine 
Journal of Science 






Published by the Bureau of Science of the Government of the Philippine Islands 

Elmer D. Merrill, M.S., Editor 

R. 0. McGregor, A.B., Associate Editor 

Albert H. Wells, A.B.; Granville A. Perons, Ph.D.; A. P. West, Ph.D. 
T. Dar Juan, A.B., Phar.D.; F. Agcaoili, A.B.; A. S. Arguelles, B.S. 

Albert E. W. Kino 


Warren D. Sbuth, Ph.D.; Roy E. Dickbrson, Ph.D. 


H. W. Wade, M.D.; Otto Sch5bl, M.D. 

F. G. Haughwout; Stanton Youngberg, D.V.M. 

Experimental Medicine 

LiBORio Gomez, M.D., Ph.D.; F. Calderon, B.A., L.M. 

Vicente db Jesus, M.D. 

Clinical Medicine 

W. H. Brown, Ph.D.; C. F. Baker, M.A.; Atherton Lee, M.S. 

0. A. Reinking, B.S.A., M.S.; L. M. Guerrero, Phar.D. 


Albert C. Herre, Ph.D.; C. F. Baker, M.A.; S. F. Light, M.A. 

C. S. Banks, M.A.; L. D. Wharton, M.A.; W. Schultze 


H. 0. Beyer, M.A.; Otto Johns Scheerer, M.A. 

Anna B. Banyea, Copy Editor 

Manuscript intended for publication and books for review should be sent 
to the editor. One hundred separates of papers published in the Journal 
are furnished to authors without charge. Additional copies may be had at 
authors* expense. 

Publications sent in exchange for the Philippine Journal of Science 
should be addressed: Library, Bureau of Science, Manila, P. L 

The Journal is issued twelve times a year. The subscription price is 5 
dollars, United States currency, per year. Single numbers, 50 cents each. 

Subscriptions may be sent to the Business Manager, Philippine Jour- 
nal of Science, Bureau of Science, Manila, P. I*, or to any of the agents 
listed on the third page of this cover. 

The Philippine 
Journal of Science 

VOL. 19 OCTOBER, 1921 No. 4 



By 0. A. Reinking ^ and G. W. Groff ' 


Siam has long been noted for its production of delicious pum- 
melos. These fruits find their way to many of the larger port 
cities of Asia, where they can often be purchased in open 
markets under the name of Bangkok or Siam pummelos. Many 
varieties are produced both for home consumption and for ex- 
port. Residents of Bangkok and other parts of Siam refer 
to the so-called seedless Nakorn Chaisri or Kao Pan pummelo as 
the best. 

Of late years the Kao Pan seedless pummelo, from Nakorn 
Chaisri, has created a good deal of interest in the United States, 
due primarily to the investigations and enthusiasm of Mr. Walter 
T. Swingle, of the Division of Crop Physiology and Breeding 
Investigations of the Bureau of Plant Industry, Washington, 
D. C. During the past eight years attempts have been made 
to introduce this form and other Siamese pummelos into the 
United States and the Philippines. Until the present investi- 
gation, however, no successful introduction of the true seedless 
form seems to have been made or, if made, the trees upon 

* Professor of plant pathology and plant pathologist of the College of 
Agriculture and Agricultural Experiment Station, Los Baiios, Laguna, 
Philippine Islands; and collaborator, Crop Physiology and Breeding In- 
vestigations, United States Department of Agriculture, Washington, D. C. 

^ Professor of horticulture and director of agricultural work, Canton 
Christian College, Canton, China, and field assistant, Crop Physiology and 
Breeding Investigations, United States Department of Agriculture, Wash- 
ington, D. C. 

181287 389 

390 ^'/ie Philippine Journal of Science 1921 

coming into bearing under the new conditions of growth have 
produced a fruit poor in quality and abundant in seeds. 

The present studies show that the **Siam seedless pummelo" 
reported in 1913 ^ as having been successfully introduced into 
the Philippines, and later in 1917 * as having produced fruits, 
was not the true seedless variety. The fruit described and 
illustrated in 1917 is of the best commercial type of Siamese 
pummelo, the Kao Phuang. It frequently happens that the Kao 
Phuang and other varieties produce a few fruits without seeds, 
but there is only one type that is generally recognized as being 
the seedless one. A direct result of the investigations has been 
our successful introduction of bud wood of the true Nakorn 
Chaisri seedless pummelo, into both the Philippine Islands and 
the United States. A small tree of the variety also was taken 
to the Philippines. 

The investigations were made during the months of June and 
July, 1920, through the cooperation of the Division of Crop 
Physiology and Breeding Investigations of the Bureau of Plant 
Industry, Washington, D. C, with the College of Agriculture 
of the University of the Philippines, Los Bafios, Philippine Is- 
lands, and the Canton Christian College, Canton, China. The 
primary objects of the investigations were: To obtain a first- 
class series of Siamese pummelos ; to make a complete study of 
the real seedless pummelo; to determine the cultural methods 
by which the proprietors of the Siamese seedless pummelo or- 
chards produce seedless fruits; to obtain varieties resistant to 
canker; to make a study of the plant diseases and insects at- 
tacking the trees ; and to look into the salting and cultural prac- 
tices in relation to quality and seedlessness. 

The low-lying region of Nakorn Chaisri is well situated for 
the production of citrus fruits, and the inhabitants of this and 
other regions have a large number of recognized varieties. 
Their pummelos can readily be classified into two main types; 
the round and the elongated. The most characteristic and best 
fruit of the first class is the so-called seedless or Kao Pan. This 
is the fruit usually referred to when residents of Siam speak 
of the delicious Nakorn Chaisri pummelo. The second class is 
represented by the Kao Phuang, a pear-shaped, elongated fruit 
produced in the Dao Kanong region. The latter is not so 

• Wester, P. J., Citriculture in the Philippines, Bull. Philip. Bur. Agri. 
27 (1913) 1-71. 

* Wester, P. J., New or noteworthy tropical fruits in the Philippines, 
Philip. Agri. Rev. 10 (1917) 21-22. 

19, 4 Reinking and Groff: Siamese Seedless Pummelo 391 

highly appreciated in Siam and is grown chiefly for export. 
The Kao Pan pummelo of Nakorn Chaisri is the more delicious 
of the two and is the favorite within Siam. While a study and 
collection of almost all of the important varieties of pummelos 
were made, we have confined the present paper to the true Sia- 
mese seedless pummelo, the Nakorn Chaisri or Kao Pan, grown 
in Ban Mai, Sarm Prarm, Nakorn Chaisri. Ban Mai is the 
only locality noted for its production of seedless pummelos and 
for the excellent flavor of the fruits. 

The studies were made primarily on the plantation of Nang 
Nui and her husband, Nai Ha, and were greatly facilitated by 
the interest shown in the work by the Department of Agricul- 
ture, Ministry of Lands and Agriculture of Siam. We are 
indebted to Mr. W. A. Graham, adviser to the Ministry; to 
Phya Manopakor, secretary to the Minister of Agriculture; to 
Phra Kasetra Raksha, director of Agriculture; and to Luang 
Bhojakara, of the experiment station. The latter accompanied 
us on our investigational trips in the fruit sections and was most 
cordial and helpful as interpreter and general informant. The 
Governor of Tachin, Phya Sakoen Kanabhirako, also assisted 
by obtaining permission to make the studies on the plantations 
and sending a guide in order that no difficulty would be en- 
countered in locating the best orchards. Besides these govern- 
ment officials, Dr. Yia S. Sanitwongse, an enthusiastic land owner 
and planter, aided by giving valuable information in regard to 
the location, history, and the general characters of the fruit. 
Without his help the study could not have been accomplished 
in such a short period of time. 


In Nakorn Chaisri the growers clearly recognize the Kao Pan 
as a distinct variety. They can describe the chief characteris- 
tics of both fruit and tree. Their claim, well substantiated 
throughout the country, is that, when grown under conditions 
other than those found in Nakorn Chaisri, the fruit soon loses 
its delicious qualities. The variety is widely known throughout 
the region, though a few growers along Tachin River are 
especially famous for their production of the Kao Pan pummelo. 

The people of this part of Siam have been greatly influenced 
by the Chinese. Chinese from southern China have settled in 
the country for generations, bringing with them conspicuous 
features of the old Chinese civilization and thereby greatly in- 
fluencing the life and customs of the people, as is shown in 

392 I'h^ Philippine Journal of Science 1921 

their manner of life, the Chinese inscriptions above their door- 
posts, and the type of citrus culture practiced by them. That 
there has been some interchange of citrus fruits between the 
two countries during past centuries is clearly evident, though 
this renowned Kao Pan pummelo is unquestionably of Nakorn 
Chaisri origin. The Chinese, through their advanced knowl- 
edge of cultural methods, have doubtless helped to develop and 
establish the strains that are to-day attracting the attention of 
the world. 

As is often the custom in Siam, much of the business, espe- 
cially in country districts, is managed by the women. One of 
these woman orchardists stated that this particular fruit had 
been known in Nakorn Chaisri for only two or three generations. 
Her mother had claimed that not more; than one hundred years 
ago only a few trees grew on their side of Tachin Rivet. This 
woman's family had grown the fruit for about fifty years and, 
about twenty years ago, had cleared the nipa swamp where their 
present grove is located. 

As a distinct variety, the Kao Pan may therefore be of com- 
paratively recent times. Other varieties grown in different 
districts, especially one known as the Koon Non, approximate 
the Kao Pan in general characteristics, but are not so seedless 
or of so fine a quality and flavor as is this delicious variety de- 
veloped under the ideal conditions for citrus culture found in 
Nakorn Chaisri. 


The Kao Pan pummelo is grown and produced in its best 
seedless and most excellently flavored form only in Ban Mai, 
Sarm Prarm, Nakorn Chaisri. The Ban Mai section is reported 
to have the best orchards in Nakorn Chaisri, and it is also stated 
that the plantations are most noted, being the ones to which 
the king and the queen mother send for their choice fruits. 

The section is on Tachin River, 30 to 40 kilometers from the 
mouth, as is shown in fig. 1. It is located southwest of Bangkok 
and can be reached by traveling on a train from Bangkok to 
the town of Tachin, a distance of between 40 and 50 kilometers, 
and then by launch up the river for about 30 kilometers. The 
region can also be reached by boat, passing through the Pasi 
Cherern Canal from Bangkok to Tachin River and then up the 
river for about 12 kilometers. 

19, 4 

Reinking and Groff: Siamese Seedless Pummelo 393 

Fig. 1. Location of the pummelo orchards at Ban Mai on Tachin Rivet ; scale about 1 to 


Some of the plantations are situated on the banks of the river, 
the trees being planted about 7 to 10 meters from the river bank 
and separated from the river by a dike. During June at high 
tide the water rises to within 30 centimejbers of the top of the 
dike; at low tide it sinks to about 2 or 2.5 meters below the top 
of the dike. Water for irrigation is taken directly from the river 
in such cases. Other plantings may be situated farther from the 
river, those studied being some 100 to 150 meters removed. 

Tachin River was at one time a branch of the Menam Chao 
Phaya, but in recent years the source of the river has been 
almost entirely cut off from the main river. The severance 
of the connection, because of lack of a large supply of fresh 

394 Tf^^ Philippine Journal of Science 1921 

water from its source, had a great influence upon the salty na- 
ture of the Tachin. According to Graham,^ 

From Paknampoh to the sea, 140 miles as the crow flies, the waters of the 
Menam Chao Phaya follow a number of tortuous courses. At Chainat, 
about 35 miles below Paknampoh, it throws off on the west side a branch 
known as the Suphan or Tachin river, which flows parallel with the 
parent stream to the sea. 

The Nakorn Chaisri district is in the southern part of the 
central division of Siam. This division is characterized by being 
a vast plain of about 14,245,000 hectares (55,000 square miles), 
stretching from the mountainous border of Burma and extending 
eastward to the high ridges marking the boundary of Cambodia. 
The plain extends southward to the Gulf of Siam. According 
to Graham,^ 

The plain lies at a very slight elevation above the sea and is subject to 
regular annual river floods which, by the deposition of vast quantities of 
silt, are slowly raising the general level. The whole area has a gentle 
slope downwards from north to south and the land falls slightly away 
at right angles to the banks of the rivers which flow on slight ridges 
of their own alluvial accumulation. There is abundant evidence that 
within recent geological times the sea flowed over a great part of the plain 
and even now the northern shores of the gulf are advancing seawards at 
the surprising rate of almost a foot a year. 

The Nakorn Chaisri section is in the delta region of the Me- 
nam Chao Phaya. In this region, according to Graham, 

The waters usually begin to rise in May and continue to do so until 
about the end of October, when the river is in full flood. Subsidence is 
gradual and the lowest level is reached about April. Sudden freshets 
and high rises are unknown in the Menam Chao Phaya. 

The river is subject to a strong tidal influence for a distance 
of 80 kilometers inland. The action of the tide is greatly in- 
fluenced by the flood time, depending upon the rainy and dry 
seasons of the year. During dry weather the tidal flow extends 
far inland and in the lower parts of the river the water is 


Graham gives the climate of Siam to be as follows : ^ 

" Graham, W. A., Siam. A Handbook of Practical Commercial' and Polit- 
ical Information. London (1912) 16. 
'Op. cit. 8. 
'Op. cit. 31-32. 

19, 4 Reinking and Groff: Siamese Seedless Pummelo 395 

The temperature of Siam, though the country lies entirely within the 
northern tropic, is considerably effected by peculiar local conditions and 
therefore varies very perceptibly in different localities. On the plains of 
Central Siam, between the months of February or March and October, 
the sea wind blows from the south or southwest almost continuously, 
mitigating the heat of the days and rendering the nights comparatively 
cool. During this period, which comprises the hot and the rainy seasons, the 
temperature rarely raises above 98° fahr. or falls below 79° fahr. From 
the end of October to February, the so-called cold season, the wind blows 
from the northeast when the maximum temperature may reach 92° fahr. 
and the minimum fall as low as 54° fahr. Formerly the climate of Bangkok 
city was very similar to that of the surrounding plains but during the 
past few years a change has become noticeable. Sir John Bowring in 
his book on Siam, gives statistics of the temperature of Bangkok over 
the period 184Q to 1847, during which the maximum temperature regis- 
tered was 97° fahr. and the minimum 54° fahr. Observations of a 
much later date give results very similar to the above but the statistics 
of the last ten years or so show an almost continual increase in the average 
daily range and at the present time, while the minimum temperature remains 
much as it used to be, the maximum reaches 105° or 106° fahr. each year 
during the hot weather and 100° fahr. during almost every month of the 
other seasons. The causes of this climatic change have not hitherto 
been explained but it is possible that they may be found in increase of 
population, in the substitution of bricks and tiles for timber and thatch as 
house building materials, or in the draining of marshes in and around 
the city. 

In Central, Northern, and Eastern Siam there are three distinct sea- 
sons, the hot weather, the rains, and the cold weather. The first extends 
from February, or March to May, the second from June to October, and 
the third covers the remaining four months of the year. When the 
northeast winds blow strong, the cold weather is very marked and, though 
the actual temperature is not below the average summer heat of Europe, 
causes some inconvenience to the people of the country. At times, however, 
the seasonal winds fail and when this happens the cold weather is 
scarcely to be distinguished from the hot. 

The statistical yearbook of the Kingdom of Siam for 1919 ® 
gives the following monthly meteorological reports, Tables 1, 2, 
and 3, on the mean temperature, the mean daily range, and the 
mean solar radiation, for ten years, from 1909 to 1918. These 
reports for Bangkok will give some indication of the temperature 
in the Naljorn Chaisri region, even though this region is some- 
what removed. 

^ Statistical year book of the Kingdom of Siam, 1919. Fourth number. 
Published by the Department of Commerce and Statistics. Ministry of 
Finance. English Edition. December, 1919. 

396 The Philippine Journal of Science 1921 

Table 1. — Mean temperature for ten years, 1909-1918, at Bangkok, Siam, 

January 25.2 

February ^ 27.4 

March 29.1 

April 30.7 

May 29.8 

June 29.1 

July 28.8 

August 28.5 

September 28.1 

October 27.7 

November 26.6 

December 25.1 

Mean for ten years 28.0 

Table 2. — Mean daily range of temperature for ten years, 1909-191^8, at 

Bangkok, Siam. 


January 12.2 

February 11.2 

March 10.3 

April 10.3 

May 9.7 

June 8.6 

July 8.2 

August 8.1 

September 7.7 

October 7.4 

November 7.9 

December , 9.9 

Mean for ten years 9.3 

Table 3. — Mean solar radiation for ten years, 1909-1918, at Bangkok, Siam.. 


January 58.9 

February 61.3 

March 63.7 

April 65.7 

May • 65.7 

June 64.6 

July 65.9 

August 63.8 

September 63.3 

October 62.1 

November 60.4 

December 59.1 

Mean for ten years 62.7 

19, 4 

Reinking and Groff: Siamese Seedless Pummelo 397 

Table 4, showing the average daily temperatures during 1919 
was submitted by the Ministry of Lands and Agriculture of 
Siam. It is for Rajburi, which is west of the Nakorn Chaisri 

Table 4. — Average daily temperature for 1919, Rajburi Province, Mantoln 

Rajburif Siam. 


January 24.4 

February 28.5 

March 3O.6 

April 32.1 

May 30.9 

June 29.9 

July 29.4 

Au^st 28.9 

September 27.4 

October 28.8 

November 27.7 

December 26.9 

Mean for ten years 28.7 

From these reports it can readily be seen that the range in 
temperature, for the years indicated, was not great either for 
the region about Bangkok or in Rajburi Province, Montoln 
Rajburi. The lowest temperature in Bangkok was 25.1° C. 
in December, and the highest was 30.7° C. in April. The range 
in Rajburi Province was approximately the same, the lowest 
being 24.4° C. in January, and the highest 32.1° C. in April. 
Since the Nakorn Chaisri pummelo region is located between 
these two stations, it is highly probable that the temperature 
for that locality falls between the figures given for the two 


Graham reports on the rainfall as follows : ^ 

The rainfall of Siam varies a good deal in the' different parts of the 
country. In Southern Siam and on the Chantaburi coast the' average is 
not far short of 100 inches for the year; in Northern Siam it is about 60 
inches, and in the neighbourhood of Bangkok about 50 inches. Until a 
few years ago the government collected no rainfall statistics but records 
have long been kept at the consulates, and by business firms and private 

'Graham, W. A., Siam. A Handbook of Practical Commercial and Po- 
litical Information. London (1912) 33-34. 

398 ^^^ Philippine Journal of Science 1921 

individuals, and these having been carefully collected and tabulated by the 
Royal Irrigation Department, and compared with the regular statistics 
of recent years, give what appears to be a fairly correct average over a 
long period. 

The comparative smallness of the rainfall in Central Siam is undoubtedly 
due to the influence of the great western mountain ranges which gather 
the clouds of the southwest or rain-bearing monsoon, and cause the precip- 
itation on their summits and slopes of the greater part of the rain which 
would otherwise be distributed more equally over the whole country. 
The rainfall is not entirely confined to the wet season, for in the neighbour- 
hood of Bangkok showers fall at intervals during the cold and the hot sea- 
sons, while towards the west and in Southern Siam the fall amounts some- 
times to several inches during the hot weather months. Snow never falls 
anywhere in Siam, not even upon the highest mountain peaks of the north 
but hailstorms, though of very rare occurrence, are not altogether unknown. 
The beginning of the wet season is usually heralded by a series of severe 
squalls and thunder-storms accompanied by heavy rain, which sweep 
down from the western heights and sometimes cause damage to property 
on the plains. During the months of September and October, heavy gales 
almost of cyclonic violence are met with in the gulf, but accidents to the 
shipping constantly plying there to* and from Bangkok are very rare. 
Waterspouts are occasionally seen both at sea and over the flooded marshes 
of the plains. 

Tables 5, 6, 7, and 8, showing the mean rainfall, the mean 
of the number of days in which rain fell, and the mean rain- 
fall on the delta of Chao Phaya River at Samutsakorn and 
Supanburi in Nakorn Chaisri Province, obtained from the 
statistical yearbook, give some idea of the rainfall in the re- 
gions near Ban Mai, Nakorn Chaisri. 

Table 5. — Mean rainfall for ten years, 1909-1918, at Bangkok, Siam, 


January 10.2 

February 16.9 

March 65.1 

April 30.0 

May 160.0 

June 150.7 

July 158.2 

August 206.2 

September 305.0 

October 241.4 

November 76.8 

December 11«8 

Mean for ten years 1,433.0 

19, 4 Reinking and Groff: Siamese Seedless Pummelo 399 

Table 6. — Number of days on which rain fell, mean for ten years, 
1909-1918, at Bangkok, Siam, 


January 1.6 

February 2.7 

March 6.2 

April 5.0 

May 16.5 

June 18.9 

July 22.0 

August 22.8 

September 22.1 

October 18.6 

November 8.0 

December 3.4 

Mean for ten years 147.0 

Table 7. — Mean rainfall on the delta of Chao Phaya River, Nakorn 
Chaisri Province, Samutsakorn, Siam, 1905-6 and 1912-13, 

January 3.7 

February 19.7 

March 49.2 

April 31.6 

May 131.3 

June 123.2 

July 105.9 

August 129.9 

September 246.5 

October 229.6 

November 70.6 

December 13.5 

Yearly average 1,154.7 

Table 8. — Mean rainfall on the delta of Chao Phaya River, Nakorn 
Chaisri Province, Supanburi, Siam, 1905-6 and 1912-13. 

* mm. 

January 15.5 

February 22.0 

March 19.0 

April 37.3 

May 164.6 

June 77.8 

July 177.3 

August 157.7 

September 294.7 

October 257.2 

November 63.1 

December 0.1 

Yearly average 1,286.3 

400 ^^^ Philippine Journal of Science 1921 

Table 9, on the rainfall for 1919, submitted by the Ministry 
of Lands and Agriculture, for Nakorn Pratom, which is north 
of Ban Mai, gives a more comprehensive idea of the rainfall 
in a region near Ban Mai. 

Table 9. — Monthly report on rainfall, 1919, Nakorn Pratom, Siam. 


January 0.0 

February 180.4 

March 26.9 

April 148.4 

May 351.4 

June 111.6 

July 199.6 

August 197.9 

September 348.7 

October 90.2 

November 79.2 

December 0.0 

Total rainfall 1,690.0 

As indicated in the preceding tables, the average total rain- 
fall in Bangkok for ten years was 1,433 millimeters. Rain 
fell during the same period on one hundred forty-seven days 
during the year. In Nakorn Chaisri Province, at Samutsakorn 
and Supanburi, which are north of the Ban Mai region, the aver- 
age rainfall during the years indicated was 1,154.7 and 1,286.3 
millimeters, respectively. The total rainfall for 1919 at Nakorn 
Pratom, which is north of Ban Mai, was 1,690 millimeters. 
The total rainfall for those districts indicated in the Nakorn 
Chaisri region varied from 1,154.7 to 1,690 millimeters. The 
rainfall at Ban Mai would undoubtedly be between these figures, 
probably being more nearly similar to those at Nakorn Pratom. 
The reports also show that a distinct dry season is present, 
which usually starts in November and continues through the 
months of December, January, February, March, and April. 


The water used for irrigation is, except during the rainy 
season, supplied entirely from Tachin River. As explained 
elsewhere, irrigation ditches are made from the river to the 
orchards. The water is rather muddy, being filled with clay 
loam, and is reported to be salty during the months between 
January or February and July. The saltness depends upon 
the beginning and the extent of the rainy season. In the 
dry season of 1920 the water was said to have been exception- 

19. 4 Reinking and Groff: Siamese Seedless Pummelo 401 

ally salty. In order to obtain an accurate idea of the salt 
content of the water, samples were collected. The water was 
collected in June, and the analyses were made in August. The 
mineral constituent such as calcium may have separated out 
during the interval, but from a sodium chloride standpoint the 
contents, after long standing, should not have altered. Sample 
1 was collected on June 7, 1920, at Ban Mai, Sarm Prarm, 
Nakorn Chaisri, from a ditch located in the center of the seed- 
less-pummelo orchard under investigation. The sample was 
taken when Tachin River was at high tide. Sample 2 was 
collected on June 7, 1920, in the same section, but from the 
canal leading from Tachin River, at the intake into the orchard, 
about 300 meters from Tachin River. The sample was taken at 
high tide. Sample 3 was taken on June 8, 1920, in the same sec- 
tion, but from Tachin River at the entrance of the canal leading 
to the orchard. This sample also was taken at high tide. The 
results of the analyses are given in Table 10. The official method 
for the determination of the chlorine was used, and the total 
chlorides were computed as sodium chloride. The values given 
are averages of triplicate determinations. 

Table 10. — Analyses of water used for irrigation of pummelo orchards 
at Ban Mai, Nakorn Chaisri, Siam." 







June 7, 1920. -. _ 

Parts per 


12, 802 


Per cent. 

Within orchard. 

Canal leading tot)rchard. 

Tachin River. 


June 8, 1920 

« Determinations by Miss H. Kenward, of the department of chemistry, College of Agricul- 
ture, Los Bafios. 

As shown by Table 10 the water used for irrigation was ex- 
tremely salty on June 7 and 8, 1920. The average sea water 
has approximately 3.5 per cent of solids in solution. Of this 
amount 2.7 per cent is reported as sodium chloride. The water 
in Tachin River at high tide at Ban Mai, Nakorn Chaisri, ap- 
proaches sea water in salt content. The analyses of the three 
samples taken at different points agree very closely. 

In order to get a comparison of the water used for irrigation 
in other pummelo sections of Siam, a sample was taken from the 
Menam Chao Phaya at Bang Bakok, near Bangkok. The Bang 
Bakok section is famous for its production of, chiefly, the Kao 

402 The Philippine Journal of Science 1921 

Phuang variety of pummelo, but other varieties of Kao pum- 
melos are also grown. The Nakom Chaisri seedless, or Kao 
Pan, variety when grown at Bang Bakok does not produce 
fruits of the excellent flavor of those grown in Ban Mai, on 
Tachin River. The seedy character of the fruits is also re- 
ported to be greater when grown at Bang Bakok. The water at 
Bang Bakok was said to be salty from about February to May, 
depending upon seasonal conditions. Usually the water becomes 
fresh in June when the rains begin. A sample of water taken on 
June 14, 1920, from Menam Chao Phaya River, at Bang Bakok, at 
the entrance of a canal leading to a Kao Phuang orchard was used 
for analysis. The sample was taken at medium tide. The 
same method for analysis was used as that described for Tachin 
River water. The total chlorine was 283 parts per million, and 
the sodium chloride, 0.04 per cent. 

The water, as shown by the analysis, used for irrigation of 
the Kao Phuang orchards at Bang Bakok, in June, contained 
only an amount of salt normal for river water. 

The results obtained show that there is a great difference 
in the salt content of water used for irrigation, at least during 
the first part of June, in the two sections. This would indicate 
that the salt content of the two rivers is different throughout the 
year and that the salt has some influence on the quality and pos- 
sibly the seediness of the fruits. Ban Mai is situated on Tachin 
River about 40 kilometers from the Gulf of Siam. The people 
reported that they can drink the river water for only six 
months of the year, as it is salty from January to June. At 
Tachin, which is 5 to 6 kilometers from the gulf, the people can 
drink the water for only about two months of the year. Sixty 
kilometers up the river from the gulf the people report that they 
drink the water for eight or nine months. These reports would 
indicate that the river is salty for at least six months of the 
year at Ban Mai ; and, according to the analysis, it is extremely 
salty even during June, when it is supposed to begin to freshen. 
A further study of the salty nature of the water should be 
conducted to determine the exact time throughout the year dur- 
ing which the water contains salt. Such a study should be 
made of both regions! ; one at Ban Mai, on Tachin River, and the 
other at Bang Bakok, on the Menam Chao Phaya. In this way 
a comparison could be made of the salt content of irrigation 
water used in a section that produces a seedless fruit of excellent 
quality with that of the water used in a section that produces 
the same kind of fruit, but of a somewhat inferior quality. Be- 

19, 4 Reinking and Groff: Siamese Seedless Pummelo 403 

fore any definite statement can be made in regard to the effect of 
salt on the quality of fruit produced, careful control experiments 
will have to be conducted. Enough evidence has been collected, 
however, to indicate that the salt water does have some relation 
to the quality of the fruit. 


In general, the surface soil in the delta region is heavy, clayey, 
and of an entirely alluvial formation. The subsoil is a heavy, 
gray, brick clay, which is rather impervious to water. The 
topsoil in the orchards at Ban Mai is a dark gray to grayish 
brown (dark brown when wet) , clay to clay loam soil. Some grit 
and sand are present. The subsoil for the same regiofi is a very 
sticky, light slate brown clay. The canal mud, which is com- 
monly used as a fertilizer, as it is composed of the decomposition 
products of organic material and the deposit of fine clay that has 
fallen into the canal, is light brown, with a slaty tint, clay to 
sandy clay, with some grit. The topsoil, collected in the re- 
gion of Bang Bakok near Bangkok, which is noted for its 
production of the commercial Kao Phuang variety of pummelo, 
is not quite so clayey as the Ban Mai soil, being a light brown 
clay to sandy clay grit. The canal mud from the same region 
is a dark gray to light brown, heavy, clay soil. Samples of the 
topsoil, subsoil, and canal mud used for fertilization were col- 
lected for physical analysis from the Ban Mai and Bang Bakok 
regions. Samples 1 to 3 are from a Nakom Chaisri orchard; 
and samples 4 and 5 are from a Bang Bakok orchard. The 
standard United States Bureau of Soils centrifuge method was 
used, and the results are given in Table 11. 

Table 11. — Physical analyses of soils in pummelo orchards at Ban Mai 
and at Bang Bakok, Siam/ 






1 .._ . . . 

Per cent. 

Per cent. 

Per cent. 

Topsoil. Ban Mai. 
Subsoil, Ban Mai. 
Canal mud. Ban Mai. 
Topsoil, Bang Bakok. 
Canal mud. Bang Bakok. 



4 . 


a Determinations by Miss H. Kenward, of the department of chemistry. College of 
Agriculture, Los Banos. 

The results of the physical analyses show that all of the soils 
are extremely clayey, containing over 50 per cent of clay. The 


The Philippine Journal of Science 


subsoil contains a larger percentage of clay than does the top- 
soil. Little sand or coarser material was present in any of the 
samples. The soil in the Bang Bakok orchard is slightly coarser 
than that of the Nakorn Chaisri orchard. 

Since the irrigation water used at Ban Mai is extremely 
salty, it would be expected that the orchard soil would also 
contain a large percentage of salt. In order to prove this point 
the soils of the various types from the Ban Mai region and, for 
comparison, those from the Bang Bakok district, where the 
water used for irrigation according to the analysis is not salty 
at that time of the year, were analyzed for their salt content. 
The follow4ng samples were analyzed from an orchard at Ban 
Mai : No. 1, topsoil; No. 2, subsoil; and No. 3, canal mud, obtained 
from the bottom of a canal leading from Tachin River to the 
orchard. The samples of soil analyzed from Bang Bakok, near 
Bangkok were from the following locations: No. 4, topsoil in 
an orchard ; No. 5, canal mud from a canal in the orchard, lead- 
ing from the Menam Chao Phaya. The soils were heated in an 
oven until weight was constant. Five-gram samples were taken, 
in triplicate. These samples were treated with water at 60° C. 
for three days, and the extract was then tested for chlorine 
by the silver nitrate titration method. Table 12 gives the results 
of the determinations. The parts per million of chlorine were 
determined from the water-soluble extract of the soils. From 
this was computed the percentage of sodium chloride present 
in the soils. 

Table 12. — Analysis of the water-soluble extract of soils in pummelo 
orchards at Ban Mai and at Bang Bakok, Siam.* 



June?, 1920- 



June 14, 1920 . 



Parts per 

Per cent. 





15. 100 







Topsoil, Ban Mai. 
Subsoil, Ban Mai. 
Canal mud, Ban Mai. 
Topsoil, Bang Bakok. 
Canal mud. Bang Bakok. 

a Determinations by Miss H. Kenward, of the department of chemistry, College of 
Agriculture, Los Bafios. 

The results given in Table 12 show that the topsoil, subsoil, 
and canal mud from Ban Mai were, all, extremely high in salt 
content. This is what would be expected, as the water used 
for irrigation was at that time very salty, as indicated in Table 

19, 4 Reinking and Groff: Siamese Seedless Pummelo 405- 

10. The topsoil, which was collected under a tree in the orchard, 
proved to contain 1.006 per cent of salt; the clay subsoil con- 
tained 0.8 per cent of salt; and the canal mud, 2.5 per cent 
of salt. The canal mud was collected from the bottom of a 
canal, the water of which contained 2.11 per cent of salt, as shown 
in Table 10. The soil in orchards at Bang Bakok contained 
only a small amount of salt at the time of the year the samples 
were taken. The topsoil contained only 0.18 per cent of salt 
and the canal mud only 0.33. During June the water used for 
irrigation at Bang Bakok, as shown by analysis, contained 
only an amount of salt normal for river water; consequently 
the orchard soil and the canal mud would not be expected to 
contain a high percentage of salt. These comparative tests 
would seem to indicate that the difference in the salt content 
of soils in Ban Mai and soils in Bang Bakok has some bearing on 
the difference in the quality of the fruit produced in each region. 
As before stated, it has been the general belief that salt has a 
direct relation to the quality of the fruit produced. Tests on 
the salting of trees would seem to confirm this belief. The 
chemical tests of the water and the soil, at least for the month 
of June, give an accurate indication as to the presence of an 
excessive amount of salt in the soil that produces the best-quality 
pummelos, and would indicate that the salt content has some 
relation to the quality of the fruit produced. 


The type of citrus culture used in Nakorn Chaisri is exceed- 
ingly well adapted to the low, wet, mud flats and nipa swamps 
of the region. From the description of the general physical 
aspect of the country it is evident that unique treatment is 
required before any measure of success can be attained. The 
methods followed are very largely those employed by the Chinese 
in the successful planting of fruit in the flood-swept delta regions 
of southern China. This type of fruit culture, almost unknown 
in the West, deserves careful consideration as of possible use 
in the development of many of the otherwise useless swamp 
areas of other parts of the world. 

Nakorn Chaisri is irrigated, drained, and sometimes flooded 
by the waters of Tachin River. Untouched by the hand of 
man, this swamp area is covered with nipa palm, Nipa fruticans 
Wurmb (Plate 1, fig. 1), or other plants adapted especially to 
wet and salty conditions. Much of the foreshore of Tachin 
River is still covered with this wild growth, giving the impres- 

181287 2 

406 ^^^ Philippine Journal of Science 1021 

sion, when viewed from the stream, of undeveloped regions be- 
yond. Occasionally the monotony of the view is broken by the 
thatched hut of some settler, probably a Chinese storekeeper 
catering to the needs of the people who are opening this watery 
jungle and very rapidly developing it. 


The first step necessary to conquer this fertile area for pro- 
duction is to provide drainage. For this the natural arteries 
leading into the main streams have been insufficient. Artificial 
channels have therefore been dug, until the whole area is now 
a network of small canals and waterways (Plate 1, fig. 2). 
These not only serve for drainage, but also provide a series of 
natural and artificial highways over which the inhabitants canoe, 
transporting themselves and their products to the main streams 
(Plate 1, fig. 3). The large Pasi Cherern Canal, dug within 
comparatively recent years, provides an important means of 
communication between Bangkok, the capital, and the outlying 
districts. It also serves to help drain the land, or to hold back 
the tidewaters for irrigation purposes during the dry season. 

Open waterways serving as drainage channels are of little 
value in carrying off the surplus water when the main streams 
are in flood. Dikes must therefore be provided to hold these 
waters back from the cultivated areas. Before the nipa swamp 
(Plate 2, fig. 1) is made ready for cultivation, it must first be 
encircled by a system of mud dikes thrown up along both main 
and branch streams. In clearing the land along the shore of the 
river a narrow strip, 1.5 to 3 meters wide, next to the stream, 
is left in natural growth (Plate 2, fig. 2). Beyond this strip 
a small, shallow basin, 1 to 1.5 meters in width, is dug, the earth 
from which is thrown to the inside of the plot, forming a raised 
embankment' a meter or less higher than the surface of the 
interior beds (Plate 2, fig. 2). Along the smaller waterways 
the extra strip of land and the ba3in are dispensed with, and 
the dikes are constructed next to the streams. 

The plot is then laid out for a series of raised beds and open 
ditches. The beds encircling the plot are usually parallel to 
the streams and a meter or so wider than those within the plot. 
This allows an extra strip for the raised or diked portion which 
is sometimes held in place by a row of trees, possibly the jak, 
Artocarpus Integra (Raderm.) Merr. (Artocarpus integrifolia 
Linn.) , planted along the inside ridge. On the inner side of this 
row of trees, the ridge portion gradually slopes off to the level 

19. 4 Reinking and Groff: Siamese Seedless Pummelo 407 

of the bed, down the middle of which at least one row of citrus 
trees is eventually planted (Plate 2, fig. 3). 

Some idea of the layout of one of these Nakorn Chaisri pum- 
melo orchards can best be had from a study of Plate 16, which 
represents an actual locality at Ban Mai along Tachin River. 
All stages of the various cultural methods are shown. In the 
lower left-hand corner is an area of virgin nipa palm and grass 
swamp in which canals have been dug from Tachin River for 
irrigation and drainage. After the canals have been dug, this 
land is prepared with ditches, as shown in the plot in the lower 
right-hand corner. The land here is set off in beds and ditches 
and has been planted to bananas, sugar cane, corn, beans, and 
peanuts. These or similar crops are grown for about five years, 
at the end of which time the beds have settled and the entire 
piece of ground is ready for the pummelo plantation. The per- 
manent pummelo plantation, with its raised beds, trees, ditches, 
and irrigation canals, is clearly shown in the upper left-hand 
corner of the sketch. All these orchards and plantings are 
drained and irrigated by ditches within each plot that lead to 
canals running to Tachin River. 

When first laid out, the interior beds are usually 5 to 5.5 
meters wide. After all natural growth has been cleared and 
burned and tree stumps have been removed, shallow ditches are 
dug (Plate 3, fig. 1). All this work is done by hand, the sticky 
soil being cut in blocks by means of a special instrument devised 
for the purpose. These blocks are then tossed upon the surface 
of the beds (Plate 3, fig. 2). The plots will not be ready for 
the planting of the citrus trees for from three to five years, as 
sufficient shade has not been provided and the soil is not in good 
tilth. At first the trenches are made very narrow and shallow; 
and the beds are often first planted to sugar cane or, sometimes, 
to peanuts, as the soil becomes quite mellow after it has been 
properly drained (Plate 3, fig. 3). Bananas are sometimes 
planted as the first crop (Plate 4, fig. 1), as these provide ideal 
shade for the young pummelo or other citrus trees which may 
be planted later. 


It will be seen at a glance that this system of canals, dikes, 
raised beds, and ditches provides not only for the drainage of 
the area, but also for its irrigation; as the water gates (Plate 4, 
fig. 2), constructed under the dikes at regular intervals, carry 
the high-tide waters of the dry season from the canals to the 

408 '^^^ Philippine Journal of Science 1921 

ditches, thus furnishing the growing plants on the beds with a 
constant supply of water. These plots are now ready for in- 
tensive cultivation and in the years to come will reveal the 
result^ obtainable on these low, othervdse useless, bottom lands 
when sufficient capital is available to clear the swamps and 
to control the waters. 


After the land is cleared and planted, some years will elapse 
before the region assumes its new aspect of a cultivated area 
(Plate 4, fig. 3). It never appears to have the regularity of a 
systematically planted western orchard, though some of the best 
Nakorn Chaisri pummelo groves approximate it (Plate 5, fig. 1). 

The dikes, first thrown up as small embankments, gruadually 
settle and are then broadened and strengthened by the fill from 
the main canals, which almost every year are deepened or 
cleaned. These eventually assume the proportions of heavy 
embankments upon which grass or cultivated vegetation soon 
takes hold. Along the main canals thatched huts are built on 
piles, often with cement steps leading down to the water (Plate 5, 
fig. 2) . Simple bridges made of bamboo, the trunk of a coconut 
tree, or perhaps a single board are placed across the stream. 

The earth at the ba^e of the dikes is often held in place by 
the systematic planting of Bruguiera sexangula (Lour.) Pers., 
the roots of which endure constant flooding (Plate 5, fig. 3). 
The slope on the stream side of the dike frequently becomes an 
impenetrable growth; but an open strip, at least 60 centimeters 
in width, is maintained along the top as an attractive and 
useful walk. On the inside of this walk is the first bed of 
fruit trees (Plate 6, fig. 1). At some suitable place under the 
dike of the first bed there is laid the hollow trunk of a palm tree, 
through which water can pass from the canal to the ditches when 
the gate is opened (Plate 4, fig. 2) . In some cases terra cotta 
pipe is used for this purpose. 

Within the diked inclosure are the beds of pummelo. These 
have become narrower and higher through the gradual process 
of widening and deepening the trenches (Plate 6, fig. 2). This 
operation of trenching is most skillfully carried out by Chinese 
laborers, as they best understand the methods whereby it is most 
efficiently done (Plate 6, fig. 3) . During the heavy rains there 
is considerable washing from along the slopes of the beds (Plate 
7, fig. 1) ; therefore, the trenches are cleaned out each year, 
and the muck is smeared back over the beds. The whole region 

19, 4 Reinking and Groff: Siamese Seedless Pummelo 409 

has thus become a labyrinth of beds and ditches across which 
one can pass with difficulty, either on narrow logs thrown across 
from bed to bed (Plate 7, fig. 2) or by hurdling the trenches. 


The pummelos are not planted for at least five years after an 
inclosure is first plotted. This allows time for the raising and 
settling of the beds, the decay of organic matter in the newly 
cleared land, and the planting of ^ome preliminary growth for 
shade. In well-planted orchards the young trees are eventually 
set out in the center of the beds with more or less irregular 
distances of from 3 to 4.5 meters between the trees. By the time 
the trees reach their most productive age, the constant deepen- 
ing and widening of the trenches has reduced the width of the 
beds to at most 4.5 meters, and the ditches between them to a 
width of from 1 to 2 meters. The depth of the trench is now 
from 1 to 2 meters from the level of the bed. 


Little scientific experimental work has been done on the effect 
of shade upon citrus plantings. That some shade is advanta- 
geous in citrus growing under tropical conditions is evident from 
the experience of native growers. In Nakom Chaisri there 
seems to be considerable intercropping, partly with the view to 
providing shade ; and numerous trees of other species are found 
planted, sometimes with uniformity and sometimes at random 
in the groves. The coconut palm (Cocos nucifera Linn.), the 
betel palm (Areca catechu Linn.) , and the banana (Musa sapien- 
tum Linn.) are the chief plants used. Betel palms are perhaps 
the most conspicuous in Nakorn Chaisri, and the trees are said 
to be very profitable. They seem to thrive under this raised-bed 
system of cultivation, and trees are often found growing along 
the edge of citrus beds (Plate 7, fig. 3). These are doubtless 
established some years before the pummelo^ are planted, and 
the center of the beds is left open for the citrus. Banana plants 
are often grown between the betel palms and the pummelos. 

In some of the other lowland regions of Siam, especially in 
Dao Kanong, a shade tree, Erythrina fusca Lour., known locally 
a,s thong lang, is grown especially to shade fruit trees and to 
sweeten and enrich the soil. Some growers have so strong a 
belief in the thong lang as to maintain that successful fruit 
culture, under the raised-bed system of irrigation and drainage. 

410 The Philippine Journal of Science 1921 

is entirely dependent upon the culture of this tree, of which 
there are said to be several varieties. The tree at maturity is 
immense, v^ith a trunk diameter at the base of 60 centimeters, 
and may attain a height of from 12 to 18 meters. The head 
shoots high above those of most fruit trees, is not very dense, 
and seems to provide a suitable amount of shade for the fruit. 
In Dao Kanong these trees are very commonly seen between 
citrus trees, but in the better Nakorn Chaisri groves they were 
not observed. The tree has an extensive fibrous root system 
which seems to thrive in wet soil, and a dense network of roots 
and rootlets extends throughout the beds and even into the 
ditches. The interesting claim of growers is that the roots of 
the fruit trees feed upon the decaying roots of these trees ; they 
also claim that the leaves fall into the trenches of water and 
decompose, and when smeared upon the surface of the beds 
provide an excellent fertilizer. 

There seems to be considerable difference of opinion among 
growers as to the extent to which intercropping and shading 
prove profitable. In the best Kao Pan pummelo groves of Na- 
korn Chaisri the growers do not make the error of excessive 
shading that often seems to be made elsewhere. In one of the 
best of these orchards it was interesting to note that, by the 
time the pummelo trees had attained the age of from ten to 
fifteen years and were nearly spread out over the surface of the 
beds, all betel palms and bananas had been removed, except the 
trees growing at the end of the beds or rows along particularly 
cherished walks extending through the heart of the pummelo 
orchard (Plate 8, fig. 1). In other groves the betel palms pre- 
dominated, even after the pummelos reached maturity (Plate 
8, fig. 2). In still another grove, betel palms were being newly 
planted between pummelos, after the latter were well grown. 
This might indicate that the betel palm was proving the more 
profitable of the two crops and was therefore supplanting the 

No accurate observation regarding the practice of shading and 
intercropping was obtainable from the growers, except that they 
consider some shade necessary for the first few years after the 
pummelo has been planted. The manager of one of the best 
groves takes out all betel palms after the pummelo trees have 
attained an age of from three to five years. In other groves 
the plantings were very much mixed, with mature banana and 
citrus trees growing side by side and with large coconut palms 

19, 4 Reinking and Groff: Siamese Seedless Pummelo 411 

growing along the borders of the dikes, casting their shade 
across the beds nearest to them. Tall betel palms, scattered 
throughout the planting, towered high above the citrus and 
banana and provided a condition of semishade which seemed 
highly advantageous (Plate 8, fig. 3). In still another grove 
twenty- to thirty-year-old pummelo trees, which apparently had 
outgrown their usefulness, were severely trimmed back and 
were being intercropped or replaced with young betel palms. 

Another form of intercropping which is commonly followed 
is to grow rice in the ditches, a row or two down either side 
of the citrus beds. The rice is set out at the beginning of the 
rainy season and not only utilizes what would be otherwise 
wasted space, but helps to keep intact the sides of the beds. 
How this practice might well be followed can be noted by observ- 
ing Plate 9, fig. 1. Doubtless these growers are unconsciously 
working out methods that should be of future value in indicating 
the lines along which scientific citrus culture should be conducted 
in the Tropics. 


Clean culture is practiced on the orchard beds of Nakorn 
Chaisri. Little difficulty is experienced with weeds, for when 
the orchards are once clean there is practically no source of 
contamination. Rarely is any effort made to maintain an earth 
mulch, but the mulch which has been cleaned out from the 
trenches and smeared over the surface of the beds is allowed 
to crack during the dry weather. The constant supply of water 
in the trenches makes unnecessary any effort to prevent evap- 
oration. During the dry season there are always at least 30 
centimeters of water in these trenches, whereas during the wet 
weather the water will often rise to within 30 centimeters of 
the surface of the beds. During seasons of exceptional floods 
the water has been known to flow over the dikes and to cover 
the beds to a depth of 60 centimeters. All growers recognize 
the ill effects of these floods, especially when the roots are sub- 
merged for too long a period. During recent years the citrus 
industry has suffered very severely from floods. This system 
guarantees the proper amount of moisture for the roots of the 
trees and shows how remarkably well the waters are controlled. 
About the only implements of culture are the specially constructed 
Chinese spade, used for removing earth from the ditches; and 
a two-pronged hoe, which is sometimes used for weeding 
the beds. 

412 ^^^ Philippine Journal of Science 1921 


A most interesting system of soil conservation and fertiliz- 
ation is practiced by the growers. This consists of cleaning out 
from the trenches all organic matter and clay that may have 
rotted and accumulated there. Dead leaves^ straw, branches, 
and other refuse naturally find their way to the bottom of these 
trenches where they decay, and by this proces;3 are made useful 
for fertilizer. All organic matter is carefully conserved, and 
one often sees under the trees (Plate 9, fig. 2) rice or sugar-cane 
leaves which, if they are not washed into the trenches, will soon 
be covered over with a coating of mud. 

Canal mud is still more fertile than the mud in the trenches, 
and a practice very common in China is followed here also. 
During the dry weather, when the canals are low, mud from 
the bottom is removed, placed along the sides of the dikes, and 
allowed to dry in blocks. After the mud is thoroughly sun- 
dried, it is broken into pieces and carried in baskets to the 
citrus trees and distributed at the rate of about two large bas- 
ketfuls to each tree. From what has already been said regard- 
ing the rise of water in the trenches, it is evident that the trees 
under this culture develop a surface root system. This fertile 
canal mud, together with the mud from the trenches, helps to 
keep these roots from exposure and to provide an important 
source of plant food. 

Nakorn Chaisri pummelo growers recognize that the quality 
of the fruit is greatly influenced, not only by soil and water 
conditions, but also by artificial fertilization. Some very inter- 
esting theories have developed regarding the effect of salt and 
paddy ash upon the seediness and what is known as the kao sam, 
or raw-rice, condition of the fruit. 

One grower in the Dao Kanong region stated that, if salt and 
paddy ash are applied to trees bearing bitter fruits or fruits 
lacking in juiciness, the fruits of the following year would be 
much less bitter and more juicy. It was also mentioned that 
the fruits would contain fewer seeds. 

The critical judge of a good pummelo in Siam will lay more 
stress upon what he calls the kao sarn, or raw-rice, condition 
of the fruit than upon seediness. There is a tendency in im- 
perfect fruit for juice sacs to harden. These are not only un- 
pleasant in the mouth but indigestible, and any fruit which 
develops this characteristic is considered inferior. Soil and 
fertilization, more than any other factors, are believed to in- 

19, 4 Reinking and Groff: Siamese Seedless Pummelo 413 

fluence this condition, and the application of paddy ash is said 
greatly to reduce the danger of its occurrence. 

Paddy ash consists of the ash of burned rice husks, of which 
a plentiful supply is always obtainable in the vicinity 'of Bangkok. 
Many large rice mills operate within this region, and the husks 
of the grain are used as fuel. At the rice mills this paddy ash 
is often dumped, because it cannot be sold or even removed 
for fertilizer. It is naturally a valuable source of potash and 
is recognized by all successful pummelo growers as an import- 
ant fertilizing element. It is usually applied to the surface of 
the beds, small piles being placed under the trees and allowed 
to work to the roots gradually. These piles can be found at the 
base of many pummelo trees in Nakorn Chaisri (Plate 9, fig. 3) . 

In addition to the above sources of fertilizer, night soil is some- 
times used. Small holes are dug in the beds under the limbs 
of the tree, and the night soil is poured in and allowed to find 
its way to the roots. Apparently this practice is not followed 
to the same extent that it is in China. 


Growers pay little attention to systematic pruning, except to 
cut out all dead branches. Shapely, low-headed trees are con- 
sidered just as important in Nakorn Chaisri (Plate 10, fig. 1) 
as they are in any western citrus grove. Trees of this shape 
are obtained not by pruning, but by the careful selection of suit- 
able branches of layering. These are so planted as to have 
the forked head only a few centimeters above the ground. The 
lower branches of the trees are then encouraged to lie almost flat 
along the surface of the bed, in many cases finally extending over 
the trench of water (Plate 10, fig. 2). In old pummelo groves 
very severe cutting back of trees is practiced. The question of 
the proper amount of pruning under tropical conditions is 
another interesting experimental field. Doubtless growers are 
erring on the safe side in not pruning too severely trees that 
have practically no period of rest. 


Western methods of propagating citrus have never been in- 
troduced into Siam, and the methods followed are almost en- 
tirely those used by the southern Chinese in their fruit culture. 
In Nakorn Chaisri the pummelos are multiplied by marcottage, 
or Chinese air layering. In this way desirable parent plants 
are used for propagation and good strains of the variety are 

414 The Philippine Journal of Science 1921 

assured if selection is wisely made. By this method low-headed 
trees are also best obtained, but it has many drawbacks. 

In Nakorn Chaisri there are no nurseries where carefully 
selected strains of the best varieties of pummelo can be pur- 
chased. In Bangkok there are a few nurseries that profess to 
sell desirable stock. Each grower usually propagates his own 
trees, rarely more than he will need for his own planting. His 
theory is that the method of layering large branches is costly; 
and that from the fruiting branch, which he must sacrifice for 
his new tree, he could actually secure fruit which in one year 
alone would bring him a higher price than the new tree when 
sold as nursery stock. The average price for a layered plant 
is from 75 satang to 1 tical each, or 30 to 40 cents in United 
States currency. At this rate it is readily seen that the nursery 
business will not prove profitable so long as this system of 
propagating is followed. Herein is found an inviting field for 
modern nursery methods. 

Marcottage is practiced in June, July, and August, after the 
rains have begun. A strip of bark is removed from the branch 
chosen. After a slight callus has formed, coconut fiber or spe- 
cially prepared earth is tied about the injured part. During 
dry weather this must be kept moistened. About one hundred 
days are necessary for the roots to form. The new plant is then 
removed from its parent and set out in a nursery bed or, more 
often, is planted directly in its permanent position. Branches 
chosen for this purpose are usually very much forked, and low- 
headed, spreading trees result (Plate 10, fig. 3). All growers 
consider these more desirable than the high-headed trees. 


The growers report that the Kao Pan, like most other citrus 
trees growing in this region, flowers in most abundance about 
four times each year. The largest number of flowers appears 
in June, at the beginning of the rainy season. This lot of flowers 
matures the largest crop of fruits. Five to six months are 
required to bring the fruits to maturity, and the largest crop 
is picked in November and sold direct or placed in storage. 

Trees of this variety are said to reach their maximum pro- 
duction when about fifteen years old and to have outlived their 
period of usefulness at about thirty years. It is difficult to 
arrive at any very definite figures regarding yield, though 
judging by the number of fruits on the trees during an off 
season, the trees of this variety must be very heavy bearers. 

19, 4 Reinking and Groff: Siamese Seedless Pummelo 415 

The yield from a grove of about 400 bearing trees was, approx- 
imately, 18,000 fruits each year, which would give an average 
of 45 fruits for each tree. Of these about 10,000 were harvested 
in November, 4,000 about February, and 2,000 each in May 
and August. 


As has been noted, the Nakorn Chaisri seedless pummelo pro- 
duces flowers and fruits the year around, but there appear 
to be four seasons at which most flowers are formed. Conse- 
quently there are four main picking seasons, the largest harvest 
coming during November. The crop that is picked in November 
is graded; as there is considerable variation 'in quality and seed- 
iness. The fruits picked at the other three seasons, about three 
to four months apart, are superior in quality, but less in 
quantity, and rarely have seeds. At these times less grading 
need be done. While there are these three or four natural pick- 
ing times, harvesting is done at almost any time throughout the 
year, according to the demand. 

Except when the fruit is picked for special, immediate sale 
to buyers who come to the plantation, care is taken to harvest 
at the proper stage of development. Color and size are the 
chief points considered in picking. The main crop is picked 
just before the fruit is mature. At this stage it has attained 
its natural size, but is still somewhat green and just starting to 
turn yellow. If the fruit is allowed to mature on the tree, the 
juice sacs are apt to develop a woody, or what is known as the 
kao sarn, or raw-rice, condition mentioned by Siamese. Some 
of the growers pick the fruits according to the sound produced 
when snapped with the finger. After picking, they are placed 
in one large pile; if no buyers are present at harvest time, the 
fruits are stored. 


The storage and the care of the fruits are important factors 
in producing the best quality, as the better ones are those that 
have been stored. After storage of from one to two months 
the skin is soft, fragrant, distinctly yellow, and much wrinkled. 
The fruits can be kept in good condition for three months, if 
put in a dry place. After this long storage the much-wrinkled 
fruits often look as if they were spoiled, but the reverse is 
true; they have become juicer and have a better flavor than 
at the time of picking. They may become bitter when stored 
too long. Little or no rotting occurs during storage, except 

416 The Philippine Journal of Science 1921 

when the fruits are roughly handled. Storing is done at the 
plantations only when buyers are few during the main harvest, 
and when the fruits are used for local consumption. 

The fruits are stored in a dry place, such as bamboo-strip 
baskets, which are placed on rafters under a nipa-palm roof 
of a house or a specially constructed storage room. The bamboo 
baskets are usually about 45 to 60 centimeters in diameter and 
60 centimeters deep. A typical storage room is a bamboo house 
with a nipa roof, the floor space being 2.5 by 3.5 meters and the 
sides about 3 meters high to the eaves. It is boarded up on 
two sides and the other two sides are made of woven bamboo 
strips to allow for plenty of ventilation. No special precau- 
tions are taken in storing; the fruits are merely piled on the 
floor in a large heap. Partition boards may be used to separate 
thQ different grades. 


Before the fruit is offered for sale it is graded, according 
to size, degree of cankeredness, and seediness. The grading 
as to the flrst two qualities may or may not be strict, and depends 
primarily on the buyer. In the November harvest an attempt is 
made to separate the fruits with seeds from those without. This 
is rather difficult, but some growers claim they can distinguish 
between these qualities. Fruits that are cankered do not neces- 
sarily sell at lower prices, because of the scarcity of this par- 
ticular variety. This is especially apparent during the Chinese 
New Year, when the demand is great. 

The fruit is m}ost often carried to market in small native 
canoes or boats; middlemen come direct to the plantation to 
buy, and they then transport the fruit to market for sale. The 
selling price at the plantation varies according to the season. 
During the large harvest in November the first-grade fruit sells 
for from 12 to 15 ticals a hundred (5 to 7 dollars United States 
currency), and the second grade for from 4 to 5 ticals a hundred 
(1.75 to 2 dollars). At this time the fruit is apt to be seedy. 
At other seasons, as in June, it sells for from 20 to 25 ticals a 
hundred (8 to 10 dollars) . The price of the seedless fruit pro- 
duced in the off seasons is, therefore, 8 to 10 ticals a hundred, 
or more than that of the seeded. When the fruits are scarce, 
as in June, the Nakorn Chaisri seedless pummelo may sell on the 
Bangkok market at the rate of two fruits for 1 tical. The 
superior quality is recognized by the people, and they pay a 

19,4 Reinking and Groff: Siamese Seedless Pummelo 417 

handsome price. Little fruit is shipped out of the country, as 
the production is small and the home demand great. 

The Kao Pan pummelo is not so good for shipping as the Kao 
Phuang pummelo, which is primarily the shipping variety. The 
seedless pummelo will, however, stand shipment if picked just 
before it is mature and then properly packed. Records of pum- 
melos shipped from Bangkok to Denmark show that fruits may 
arrive in good condition after a voyage of from six to eight 
months. These fruits were individually wrapped in paper, so 
as to prevent infection at points of contact, and then packed in 
a box with slat sides to allow for ventilation. Trials with ap- 
parently good results were cited in which the fruits were sent to 
Denmark after being varnished and then packed as above. 
Fruits picked one month previous and then taken by us to Cantpn 
and the Philippine Islands arrived in good condition after having 
been packed for one and one-half months. They were wrapped in 
paper and then packed in a woven bamboo-strip basket. Fruits 
of the Kao Pan variety picked at Ban Mai, Siam, between May 
11 and May 30, 1920, wrapped individually in paper, and packed 
in a bamboo-strip basket and shipped to Los Banos, Philippine 
Islands, arrived in excellent condition and remained so until 
August 25 (Plate 13, figs. 1 and 2). If properly picked, with 
good shipping conditions and ordinary storage conditions, the 
fruits will remain in excellent shape for two or three months. 


Most of the trees of this variety are well rounded and shapely 
(Plate 10, fig. 1). Jen-year-old trees, grown from air layers, 
attain a height of 2.5 to 3 meters and a spread of 3.5 to 4.5 
meters. They are not so markedly upright as are most pummelo 
trees, but incline more to^ the rounded outline of the orange 
tree (Plate 10, fig. 1). As grown in Nakom Chaisri, the trees 
are all low-headed, branching from the surface of the ground 
into two or more forks (Plate 9, fig. 2) . As noted under prun- 
ing, this low-headed, forked characterestic is largely obtained 
by the growers through selection of branches for layering that 
are so shaped as to produce the type of tree they desire. 

The branches vary somewhat in $he angle of growth that they 
take (Plate 10, fig. 2), but in general they tend to spread rather 
than to shoot upright. Usually the young branches take an angle 

418 The Philippine Journal of Science 1921 

of about 45° from their main stem. The tree has a very dense 
growth of branches and twigs (Plate 10, fig. 2) . The branches 
are almost thornless ; when thorns do appear, they are very rudi- 
mentary and hardly noticeable. The bark of the tree, which is 
quite smooth, is brown and streaked with yellow lines. The 
young stems, somewhat downy, are of characteristic pummelo 
shape and color. 


The leaves of the trees of the Kao Pan variety form somewhat 
densely on the trees, but do not appear to be of so hardy a 
character as does the foliage of other varieties perhaps nearer the 
wild stock. Leaves vary somewhat in size and shape, being 
from 7.5 to 14 centimeters long and 4.5 to 7 centimeters wide; 
the margin is slightly wavy or scalloped, especially on the upper 
half of the leaf. Most of the leaves are of the small type and 
are flattened at the tip and markedly indented. The petiole, 
with variable wings, varies in size and shape, in many leaves 
being distinctly narrowed and shaped somewhat like that of 
the orange leaf. The veining of the leaves is very prominent 
and extends from the midrib at an angle of almost 60°. There 
is a distinct marginal veining of the leaf. 


Trees of the Kao Pan pummelo in Nakorn Chaisri flower three 
or four times each year and are practically ever-bearing. The 
flowers are usually formed on young stems of considerable length, 
which shoot out quite upright from the branches of the previous 
year's growth (Plate 11, fig. 1). At the terminal end of the 
flower stem there may or may not appear a cluster of leaves 
(Plate 11, fig. 1). The number of flowers appearing on a single 
flower stalk varies; sometimes there are as many as fifteen or 
twenty in a cluster (Plate 8, fig. 4), which measures 7.5 centi- 
meters in length and 4.5 centimeters across. There seems to be 
little difficulty for the flowers to set fruit, but as a rule not more 
than two or three fruits are allowed to mature in a cluster. The 
petals of the flowers are about 2 centimeters long and spread 
to a diameter of about 5 centimeters when the flower opens 
(Plate 11, fig. 3). The white color and the decided fragrance 
make a cluster of the flowers very attractive. 


The fruit of this variety, unlike many of the pummelos grow- 
ing in the Orient, is of a decidedly rounded, somewhat flattened 

19, 4 Reinking and Groff: Siamese Seedless Pummelo 419 

form (Plate 13, fig. 2), with no evidence of neck formed by a 
raising of the calyx end of the fruit. Nearly mature fruits 
picked from the trees in June averaged 44 centimeters in lat- 
itudinal circumference, 40 in longitudinal circumference, 11 in 
latitudinal diameter, and 10 in longitudinal diameter. They 
are very solid. The pistil end is flat and smooth and, usually, 
only very slightly depressed or cupped; it is not even slightly 
furrowed. The calyx end is also usually quite flat or sometimes 
slightly cheeked on one side ; the furrows, if any, are very short 
and shallow. 

The rind is smooth for a pummelo, being only very slightly 
roughened by the numerous small oil cells which group them- 
selves conspicuously over the surface (Plate 13, fig. 2) . These 
are rounded, about half the size of a pirihead, and grouped more 
or less uniformly, averaging about 3 millimeters apart. Un- 
like many varieties of pummelos there are no bottle-shaped oil 
cells; although those embedded within the rind are large, they 
do not extend into the rind to a depth of more than about 
1 millimeter. Oil exudes very slightly from the rind upon pres- 
sure of the thumb and fingers, and a very slight aroma is pres- 
ent in fruit which is still solid. The fruits tend to become 
somewhat shriveled when placed in storage. The softening and 
wrinkling of the rind, attended by reduction of size, detracts 
somewhat from its appearance (Plate 14, figs. 1 and 2). The 
rind is from 1 to 2 centimeters thick when the fruit is first 
picked. The thickness varies according to the maturity (Plates 
13, fig. 1; and 14, fig. 1), It is very pithy, but in storage be- 
comes somewhat reduced in thickness. The Kao Pan is a fine 
keeper, and residents of Siam purchase fruits of this variety in 
quantity and store them for months at a time, using them as 

The rind clings tightly to the sections, and considerable art 
is required in order to open the fruit in an attractive way for 
the table. The Siamese usually cut the fruit longitudinally into 
irregular pieces and then tear apart the fleshy sections, finally 
stripping off the rind (Plate 12, fig. 1). If one attempts to 
open the fruit as one would a navel orange, he will find not only 
that the rind sticks tightly to the sections, but that the sections 
adhere tightly to one another; and it is almost impossible to 
tear them apart neatly with the partition walls intact (Plate 12, 
fig. 2). When the fruit is opened in this way, the large core 
is at once evident and its tendency to stringiness is revealed. 
By cutting across a number of fruits latitudinally, almost closed 

420 ^^^ Philippine Journal of Science 1921 

cores are usually exposed (Plate 13, fig. 1). They are from 1 
to 2 centimeters in diameter and are rounded or irregularly 
shaped. In some fruits the core forms a single string, while 
in others there is an open center with the core extending down 
one side (Plate 12, fig. 3) . 

There are from twelve to fifteen locules, or sections, in a 
fruit. These are uniform in size and shape and in some fruits 
have a slight tendency to extend into one another (Plate 10, fig. 
2). The partition walls are thick and tough and usually are 
not allowed to enter the mouth, as they can be readily stripped 
from the juice sacs without bursting the latter or even tearing 
them apart. Sometimes a portion of the wall remains attached 
to the base end of the sections, and this may enter the mouth, 
producing the only rag noticeable. 

The nature of the juice sacs is a characteristic feature of this 
fruit and unquestionably helps to make it attractive. They 
average about 2.5 centimeters in length and 4 millimeters at 
their greatest diameter. The long sacs are grouped more or 
less in parallel lines, and on the base of the section they are 
joined by shorter, more rounded or irregularly formed, juice 
sacs. Each little juice sac is a unit in itself and can be readily 
separated from the partition walls and from the others without 
breaking its tender walls. Thus each section of the fruit when 
opened by the hand reveals a large mass of attractive juice sacs, 
which readily crumble apart and can be carried to the mouth in 
any quantity desired without causing any of the juice to run 
out. A little pressure of the tongue will break these sacs, re- 
leasing the delicious juice contained therein and leaving little 
trace of rag, for the partition walls of the sections have already 
been removed and therefore never reach the mouth. 

The flesh is white and very juicy, with a delightfully sweet, 
very mildly acid flavor. The bitter pummelo taste, which in 
many varieties is not altogether pleasing, is in this variety 
only slightly evident and blends very nicely with the sugars 
and acids of the juice. Those who eat the fruit for the first 
time commonly declare it to be one of the best fruits of this 
class they have ever tasted. 

As has been stated, the Kao Pan is practically an ever-bear- 
ing variety, and the quality of the fruit varies with the season 
of production. In Nakorn Chaisri there is a decided tendency 
to seedlessness during a part of the year. Growers say that 
the same trees which bear seedless fruits in June, or fruits with 

19, 4 Reinking and Groff: Siamese Seedless Pummelo 421 

very small abortive seeds., v^ill bear fruits in November which 
often have large mature seeds, frequently considerable in num- 
ber. The contention is that there are no individual trees which 
always carry seedless fruits, but fruits of the same tree vary 
greatly. The fact, that the same trees that produce seedless 
fruits at times are said to produce seedy fruits during other 
portions of the year, destroys the common theory that the Kao 
Pan is a true seedless variety. Moreover, when taken to other 
places it usually becomes seedy. There can be little question, 
however, that this variety has a greater tendency to seedless- 
ness than many others. A strictly seedless strain might be se- 
cured through an extended study of tree and bud variation. 
In most of the fruits there is always some evidence of rudimen- 
tary seeds. 


From the general description of this fruit and the tree upon 
which it grows, the likeness of some of its characteristics to 
those of the orange is apparent, though somewhat remotely 
hidden. This might possibly raise a question as to whether 
the variety might possibly be a natural hybrid of Citrus maxima 
(Burm.) Merr. {Citrus grandis Osbeck), the pummelo, and 
Citrus sinensis Osbeck, the common or sweet orange; or, at 
least, it might possibly possess some orange characters. Cer- 
tainly, the nature of the tree, the shape of the leaf and the fruit, 
and the size and the shape of the oil cells would indicate a 
tendency in this direction ; but this fruit is rightly classed as a 
variety of pummelo. 


A study of citrus diseases of Siam was made during June and 
July and later published.^^ Though the season was still dry, it is 
believed that the major diseases were encountered. The se- 
verity of these is undoubtedly much greater during the rainy 
season, and some maladies, other than those noted by us, may 
be present. 

Citrus canker is the same in general formation on the various 
species and varieties of citrus growing in Siam as on those 
growing in the Philippine Islands. Citrus maxima (Burm.) 
Merr. (C grandis Osbeck) and C. aurantifolia (Christm.) Swin- 

" Reinking, Otto A., Citrus diseases of the Philippines, Southern China, 
Indo-China and Siam, Philip. Agri. 9 (1921) 121-179. 

181287 3 

422 T'he Philippine Journal of Science 1921 

gle are severely attacked in both regions. The Nakorn Chaisri 
seedless pummelo is extremely delicate and, consequently, most 
apt to be severely attacked by canker and other diseases. An ex- 
tensive study of the several varieties may shov^ some slight resist- 
ance, but in general all trees of this type are attacked. Fruit 
rots may take place while in storage, especially on fruits that 
have been injured during picking and handling. This trouble 
can, however, be entirely avoided by proper handling and ship- 
ping. In order to obtain the highest possible yield of unmarred 
fruits of this variety, the best of culture and care will have to 
be exercised. If this be done, it will be possible to produce the 
Nakorn Chaisri pummelo commercially. An intensive study of 
this type should be undertaken to determine whether or not a 
more resistant strain can be developed. 

The diseases listed below may be found on all varieties of 
pummelos. However, the susceptibility to disease of varieties 
in the different groups seems to vary. 

Algae ; parasitic ; undetermined. Parasitic algae may produce 
a leaf spotting which, however, is not serious. 

Aschersonia aleyrodis Webber. Scale insects on citrus trees 
may be attacked by Aschersonia aleyrodis Webb. It is a bene- 
ficial fungus as it reduces the number of coccids. Scale insects 
of various species, as later listed, may be severe on stem, leaves, 
and fruit. 

Bark rot ; Diplodia. A common bark rot caused by a Diplodia 
may attack the pummelo, but does not appear to be serious. 
The same fungus produces the gummosis of trees and the black, 
sooty fruit rot. 

Canker; Pseudonfionas citri Hasse. On the Nakorn Chaisri 
seedless variety the canker is extremely serious, as it attacks 
leaf, stem, and fruit (Plate 15, fig. 2). There appears to be 
no individual variation as to susceptibility. Brownish, can- 
kered areas are produced on the aflfected parts. The growers 
do not consider the disease to be a serious one and believe that 
it is due to a caterpillar or to some other insect. If only a 
few cankered infections develop on the fruits, the planters say 
that the sale is not hindered ; but if the affection is severe, the 
price obtained for such fruits is somewhat lower. The great 
demand, especially among the Chinese, for the best types of 
fruit and their comparative scarcity account for their ready 
sale even though defaced. The growers recognize the fact that 
a rot does not start from the xankered area. The canker ap- 
pears to be more prevalent in well-kept plantations where the 

19. 4 Reinking and Groff: Siamese Seedless Pummelo 423 

trees are vigorous. In neglected orchards it is not so much in 
evidence. Counts of cankered fruits of the Nakorn Chaisri 
seedless variety in storage give some indication of the severity 
of the disease. Of 128 fruits examined, 74, or 58 per cent, were 
found to be more or less severely cankered. Young plants, 
introduced and grov^ing under Philippine conditions, at the Col- 
lege Of Agriculture, Los Bafios, are badly affected. 

Epiphytes; parasitic; Loranthits species. The most serious 
trouble on citrus trees in general is caused by Loranthus, of 
which the following species were found to attack the Siamese 
pummelo: Loranthus ferrugineus Roxb., L. 'parasiticus (Linn.) 
Merr., and L. pentandrus Linn. If the epiphyte is not cut out, 
it soon envelops the entire tree. The parasite has a wide range ; 
it is found on mangos, mangosteens, and durians, and on the 
native tree Bruguiera sexangula (Lour.) Poir. that is used to 
line dikes. Care should be taken to remove the epiphyte from 
all trees affected. 

Fruit rot; Diplodia. Fruits in storage or those that have 
been allowed to decay on the ground are frequently subject to 
a black rot caused by a Diplodia (Plate 15, fig. 1) . The trouble 
is not serious. It is not common on fruits on the tre^s or in 
storage. Rough handling in picking and storing makes the 
fruits more susceptible to rot. All black-rotted fruits should be 
collected and burned, as the fungus causing the disease is the 
same that produces bark rot and gummosis. 

Fruit rot ; Sclerotium. A Sclerotium may cause rot on fruits 
that have fallen to the ground. The disease is characterized 
by the formation of small, spherical, brownish, sclerotial bodies 
over the decayed area. 

Gummosis; Diplodia, Gummosis may be severe in neglected 
orchards. It is of two kinds, that due to a Diplodia and that 
due to malnutrition. 

Lichens; undetermined. Lichens of various sorts are found 
on trunk, branches, leaves, and fruits. Little damage is done 
by them. 

Mottled leaf; nonparasitic. A typical mottled leaf with the 
yellowed areas between the green veins is common, but is not 
considered serious. 

Russet; undetermined. Fruits may be russety. Whether 
this is due to injury caused by rubbing against branches or to 
fungi could not be determined. In some cases the disease ap- 
peared to be typical, due to the wither-tip fungus Colletotrichum 
gloeosporioides Penz. 

424 ^^^ Philippine Journal of Science 1923 

Scaly bark; undetermined. Old neglected trees are subject 
to a rotting of the bark and a consequent sloughing of the dead 
parts. The trouble is only serious in older, poorly kept or- 
chards. It is not a typical bark rot. 

Septobasidium albidum Patouillard. A thick, brown, leath- 
ery growth of Septobasidium albidum Pat. may be produced 
over stems attacked by scales. Usually no damage is done by 
the fungus. 

Sooty mold ; Meliola citricola Sydow. A black mold may de- 
velop on leaves and stems, especially in the presence of scale 
insects. The trouble is not serious. 

Witches' broom; nonparasitic. In poorly kept, improperly 
pruned orchards, witches' broom may develop on the branches. 
The trouble is not serious. 

Yellowing, or chlorotic condition; nonparasitic. During the 
extreme dry season, the trees may look sickly, being yellowed. 
This condition is commonest where irrigation is not properly 
practiced. The Nakorn Chaisri seedless pummelo is not a 
hardy, robust tree; consequently it is easily weakened by un- 
favorable conditions. Often, due to neglect, the trees are old 
at ten years of age. Well-kept orchards have healthy trees, 
free from the majority of serious plant pests, and produce fruit 
for thirty years. The age of the tree is frequently dependent 
upon the care given it and upon the extent of the dry season. 
Healthy trees produce most abundantly between eight and fifteen 
years of age. In poorly kept orchards, during the extreme dry 
season, the trees are sickly, the leaves wilt and drop, and there 
is a die-back of the branches. The trees will not stand flooding. 
The greater number of seedless pummelo trees in the Ban Mai 
section that have died were lost either through lack of attention 
or through heavy floods. The owners do not replant with the 
Kao Pan trees, because they are delicate and require a good deal 
of attention. They recognize the hardiness of the Kao Phuang 
pummelo and are planting this variety. 

On dead branches the following may develop : Schizophyllum. 
commune Fr., Tryblidiella rufula (Spreng.) Sacc, and Hetero- 
chaete tenuicula (Lev.) Pat. 


Insect attacks may cause great losses. The leaf miner, the 
flea beetle, ants, and scale insects do the most damage. The 
insects listed are those most generally found and, consequently, 
those that may cause the greatest amount of destruction. 

19, 4 Reinking and Groff: Siamese Seedless Pummelo 425 

Ant; mod dam; Dolichodems species.^^ Harmless black ants 
are often found on the leaves, especially in the vicinity of scale 
insects. They attend the scale to obtain the sweet secretions 
of the latter, but cause no damage to the tree. 

Ant; mod kan fie; Pheidologeton species. A reddish fire ant, 
so called because of its burning bite, may cause serious damage, 
even to the extent of killing some of the trees. The ant attacks 
and kills the young roots, and also is reported to attack the 
older roots, the trunk just below the ground surface, and even 
the young twigs and leaves. In some cases ants of this species 
girdle and kill the trees. Rot-producing and parasitic fungi 
frequently gain entrance to the tree through injuries caused 
by ants. Fruit touching the ground may be attacked by ants 
and other insects, thereby opening a way for fungi. 

The control consists in flooding out the ants and burning 
them. In October and November when the water is high, dur- 
ing the flood periods, the growers allow the water to flow 
over and stand on the fields for from two to three days. By 
this method the ants are forced out of the ground. They come 
to the surface and seek the highest points of land. The owner 
then travels about with a lighted torch of long grass and burns 
them. Termites, small red ants, and black ants are also killed 
in this way. 

Ant, red tree; mod dang; Oecaphylla smaragdina Fabricius. 
The large red, or cedar-colored, tree ant is common on the 
branches of trees. It does no damage to the tree directly, but 
makes nests of the living leaves. The ants of this species at- 
tend scale insects to obtain the secretions from the bodies of the 
latter, and are found throughout the East ; they are a pest dur- 
ing harvest time, and attack and bite the pickers. 

In the citrus groves about Canton it is reported that the or- 
chardists import the ants and place nests of them on the trees 
to destroy caterpillars, probably tent caterpillars, which at times 
do serious damage. The trees in orchards are connected by 
bamboo poles to enable the ants to pass from tree to tree. 

Flea beetle. Flea beetle attack on the leaves is characterized 
by the epidermal layer and the mesophyll being eaten away on 
one side of the leaf down to the epidermal layer on the other side. 
Severe injury may be done. 

" The ants listed were determined by H. E. Woodworth, College of 
Agriculture, Los Baiios. 

426 The Philippine Journal of Science 1921 

Leaf miner; Phyllocnistis citrella Stainton. The leaf miner 
may cause much loss, especially on young nursery trees. The 
miner seems to prefer the young leaves. Badly attacked young 
leaves are stunted, v^rinkled, and rolled up. On older trees 
the effect is not so noticeable. The insect may spread canker 
infections. The leaves of the various Loranthus species may 
also be attacked. 

Scale insects. Scale insects may be present in abundance 
on stem, leaves, and fruit of pummelos. The following have 
been indentified: Chrysomphalus aonidium (Linn.), C aurantii 
(Mask.), Coccus hesperidum (Linn.), Lepidosaphes gloverii 
(Pack.), Parlatoria brasiliensis sp. nov., P. ziziphus (Lucas), 
Pseudaonidia trilobitiformis (Green), and Saissetia sp.^^ ^^ 
examination of fruits in storage, showed that of 37 only 2, or 
.05 per cent, had been attacked. 

Termites; pluag. Termites do some damage by eating the 
roots. No serious losses are produced. 


The Nakorn Chaisri seedless pummelo, Kao Pan, is only pro- 
duced in its seedless form, with its excellent flavor and juiciness, 
in a restricted region on Tachin River about the section, or Tom 
Bol, of Ban Mai. If taken to regions other than Nakorn Chaisri 
it is said to deteriorate greatly, in that the fruits contain many 
seeds and the quality of the juice sacs is not so good. This 
deterioration is often attributed to lack of salt in the soil. As 
has been pointed out, the tidewaters of Tachin River in Na- 
korn Chaisri are salty for the greater part of the year, from 
January to June or July, depending upon the rains. Growers 
say that if the season of salty water is long, the fruits of the 
following year are better in flavor and have less seeds. They 
report that there is no danger, in the pummelo sections, of a 
too salty condition of the soil through this natural process of 
salting by means of the tidewater. In other regions there have 
been attempts to maintain the Nakorn Chaisri standard of per- 
fection by the artificial application of coarse sea salt. 

Some orchardists claimed that the trees grown away from 
Nakorn Chaisri had fruit without seeds ; others stated that many 
seeds developed. It was impossible for us to ascertain defi- 

" Determinations by Harold Morrison, of the United States Bureau 
of Entomology, Washington, D. C. 

19, 4 Reinking and Groff: Siamese Seedless Pummelo 427 

nitely which was the case in most instances; but the evidence 
obtained indicated that the trees produced fruits without seeds 
during certain seasons, and, at other seasons, fruits with many 
seeds. The quality of the fruits grown elsewhere was always 
much poorer. The Governor of Tachin, Phya Sakoen Kanabhi- 
rako, reported having sent one hundred marcots to various sec- 
tions of the country about Bangkok, and that the fruits grown 
from these were poor. The first crop from such trees was re- 
ported to be fairly good, but in succeeding years the fruits de- 
teriorated, becoming seedy and poor in quality. Direct evidence 
of such cases could not be thoroughly investigated; so, for the 
present, it is necessary to rely upon the evidence given by 
growers. Planters in Bang Sorn, a region above Bangkok, 
claimed that trees of the Nakom Chaisri seedless pummelo taken 
from Ban Mai and planted in their section became very seedy 
and resembled in all respects a variety grown there called the 
Koon Non. 

Various causes are given for the superior quality of the Kao 
Pan when grown in the section about Ban Mai. The theory 
that seems to be most commonly accepted by the growers is 
that the salty character of the water used for irrigation produces 
the exceptional quality of the fruit. Other possible causes im- 
mediately present themselves. The ones of most importance 
are general culture, such as the selection of a proper orchard 
site and the practice of proper irrigation, cultivation, fertiliz- 
ation, and cleanliness. The relation of seedless fruit to pollina- 
tion, soil, meteorological conditions, curing, and storing may 
be factors. 

The orchards in Ban Mai are located in reclaimed nipa-swamp 
areas. The nipa palm, Ni'pa fruticans Wurmb, grows only 
under salt conditions a^d is especially common along tidal 
streams. The orchards are all irrigated with river water, as dis- 
cussed in full under culture. The growers allow the salty river 
water to flow into the ditches between the trees, even during 
the times that it is most salty, at each high tide, which is twice 
in every twenty-four hours. Intakes and outlets are supplied 
for the orchards ; at low tide the water flows out and at high tide 
it runs in, as desired. During the extreme rainy season no river 
water is allowed to flow into the ditches. This is automatically 
regulated by the trapdoor intakes. The plantations are on land 
that at high tide is approximately 60 centimeters above the water 
and at low tide between 2 and 2.5 meters above. It can readily 
be seen that the trees are constantly, throughout the main part 

428 The Philippine Journal of Science 1921 

of the year, supplied with salty water. During November, at 
which time the crop is the heaviest, the river is reported to be 
free from salt. At this season the fruit is poorer in quality 
and is more or less seedy. 

The chemical analyses of water and water extraction of or- 
chard soil from the Nakorn Chaisri region showed that in June 
the salt content is great ; and the analysis from the Bang Bakok 
region showed only a trace of salt to be present. This would 
indicate that there is a difference in the salt content of the water 
throughout the year in both localities. The difference in quali- 
ty of the Kao Pan grown in these regions may then be due to 
the variance in the salt content of the soil and the water used 
for irrigation. According to the people at Tachin, 5 or 6 kilo- 
meters from the gulf, the fruit cannot be produced there because 
of the extreme salt content of the river water. 

The salty character of the water used for irrigation appears 
to account for the superior flavor of the Nakorn Chaisri pum- 
melo. The commercial pummelo, Kao Phuang, grown in the Dao 
Kanong region of the Menam Chao Phaya just below Bangkok, 
seems to react to the salt influence when grown in Nakorn 
Chaisri, where it produces a better quality of fruit. The flavor 
is reported to be improved also when the Kao Phuang is grown 
down Menam Chao Phaya River near the Gulf of Siam where the 
water is saltier. In these regions, the system of irrigation with 
salt water is similar to that practiced in the Nakorn Chaisri 
region, and the water of the Menam Chao Phaya would un- 
doubtedly be somewhat of the same salty character as that of 
the Tachin. 

Experiments have been conducted in various parts of Siam to 
test the relation of salt to fruit production. The salting of 
pummelo trees to produce better quality is by no means a new 
practice, nor is it restricted to Siam. Cameron ^^ cites a case 
of salting in India. Under Citrus decumana Linn, and the Can- 
ton pummelo he writes the following: 

A dressing of salt to the roots of the trees I have been told by a friend, 
who tried it upon several in his garden, has a surprising effect in improv- 
ing the quality of the fruit, renderng it tender as an orange, and all but 
bursting with juice. 

An instance of salting a Nakorn Chaisri seedless pummelo 
tree at Bangkok was given us by Dr. Yia S. Sanitwongse. The 

" Cameron, J., Firmingers Manual of gardening for India. Thacket 
Spink and Co., Calcutta (1904) 280-281. 

19. 4 Reinking and Groff: Siamese Seedless Pummelo 429 

complete history of this tree could not be obtained, but it is 
reported as being a seedless pummelo from the Ban Mai district. 
The tree in its new locality had been producing fruits that were 
seedless but not juicy, having woody juice sacs. Raw salt was 
scattered around the base of the tree. The salt application was 
reported to have produced a better-flavored and juicier fruit. 
There was no effect upon the seedy character. Other trials 
about Bangkok produced the same result. People in the north- 
em part of Siam, at Chieng Mai, also practiced the salting of 
trees and reported an improvement in juiciness and flavor. The 
seedy character in these cases was not changed, as the fruits 
always remained seedy. 

The application of salt along with rice-paddy ash is practiced. 
This method was reported in every case, after the first year's 
application, to decrease the bitterness of the fruits and to increase 
their juiciness. Some reported that the treatment had the 
effect of producing seedlessness, but no direct evidence on this 
point could be given. Two systems of application are practiced. 
The first method consists in applying the salt and rice-paddy 
ash directly to the trunk of the tree by digging the soil away 
from its base. In the second one the salt and paddy ash are 
scattered about the base of the tree, away from the trunk. Or- 
dinary sea salt is used. It is recognized that too much salt will 
kill a tree. Usually only one application is made each year. 
These methods have not been practiced on a large scale; only 
a few trees, here and there, were so treated. Without a doubt 
the salt treatment produces fruit of better quality as to flavor 
and juiciness, but there is no effect, so far as any evidence 
shows, upon its seediness. 

It has long been known that common salt, sodium chloride, has 
a marked effect on certain soils. The cause of this is not clearly 
understood. The addition of sodium and chlorine as soil con- 
stituents undoubtedly has no effect, as there is usually an excess 
of these elements in available form in soils. It is known that 
the addition of sodium chloride, especially to clay soil, liberates 
certain plant nutrients such as calcium, magnesium, potassium, 
and phosphorus. It may be that such action is responsible for 
some of the beneficial effects observed in the regions under dis- 
cussion, as the soil in those regions is primarily clayey. Soil 
structure also is improved by the addition of salt, and salt has 
the effect of conserving and distributing the soil moisture. 
Transpiration is retarded and the film movement increased by 

430 The Philippine Journal of Science 1921 

salt in solution. The movement of water is increased in this 
way ; consequently, the roots are always well supplied, especially 
in locations where the irrigation ditches are constantly filled 
with salty water. Experiments elsewhere have shown that not 
all soils ^re benefited by salt application. 

It would seem that the beneficial effect of salt in the Nakom 
Chaisri orchards is primarily due to liberation of certain plant 
nutrients and increase in the film movement whereby the roots 
are constantly kept abundantly supplied with water. The chem- 
ical effect of salt, when taken into the plant, especially in 
relation to sugar transport, may also have some bearing on the 
quality of the fruit. Further experiments would have to be 
conducted to determine these points. 

Another important point in the production of high-quality 
fruit is general culture. The Kao Pan is not an extremely hardy 
and robust tree and responds readily to culture. The orchardists 
who produce the very best fruits are those that use the most 
careful orchard practices, such as the selection of a proper or- 
chard site and the practice of proper irrigation, cultivation, 
fertilization, and cleanliness. The best orchards have been lo- 
cated in the Ban Mai region where the salty character of the 
Tachin River appears to be at its best. The growers in these 
regions have small holdings and take exceptionally good care of 
their trees. The nearness to the river does not appear to make 
any difference, as orchards located on the banks and others a hun- 
dred meters from the bank produce fruit of equal quality. The 
height of the land above the water, according to the growers, 
has some effect on the fruit ; for, if the land is a little too high 
above the water, the best pummelos cannot be developed. This 
would indicate that irrigation with a constant supply of water 
is essential. Keeping down weeds and the production of a dust 
mulch are practiced in the best plantations. The application 
of fertilizer, such as wood ashes or paddy ash and canal mud, 
at least once each year, is essential. Pruning of all dead 
branches and removal of parasites such as Loranthus must be 
done to obtain the best results. Trees in neglected orchards, 
even in the heart of Ban Mai district, are sickly and dying, 
and they produce a distinctly inferior fruit, with thicker skin and 
poorer flavor, than do neighboring orchards where the best 
attention is given. 

It appears from these investigations, that Kao Pan trees grown 
out of the Nakom Chaisri region, even when the salt require- 

19, 4 Reinking and Groff: Siamese Seedless Pummelo 431 

merits were right, have not been given proper culture and care. 
The orchardists gave the same treatment as is given to the 
Kao Phuang tree, which is hardy and can withstand lack of 
attention. Therefore, after the other essential needs such as 
salt application have been provided for, proper culture is un- 
doubtedly an important requirement for the production of a 
Kao Pan seedless pummelo of good quality. Careful experimen- 
tation in the use of proper culture will have to be conducted to 
prove whether or not the true quality and seedless characters 
can be retained when the pummelo is grown in any other than 
the Nakom Chaisri region. 

The relation of seedless fruit to pollination and the presence 
or the absence of the foreign pollen will have to be carefully inves- 
tigated before any definite statements can be made. Whether 
the pollen of the Kao Pan is not viable for the main part of the 
year, or whether the stigma is not receptive, cannot be stated. 
All that is known is that the fruit contains only abortive seeds 
in every month except November. During this month the great- 
est crop is produced, and the fruit is apt to be seedy. Wild 
bees are present in the orchards at all times, and provide a 
method for the interchange of pollen. Frequently interplanted 
with the seedless pummelos are other varieties of pummelos, 
such as the Kao Phuang, that are extremely seedy, as well as 
seedy limes, and mandarin and other oranges. In no case was 
there any indication that the seedless pummelo trees adjacent 
to the other seedy citrus trees were any different from the rest 
of the Kao Pan trees, from the standpoint of seediness. Seeds 
were never developed in the fruits of these trees except as re- 
ported, in November. From all observations it appears that the 
Kao Pan is a distinct variety, which is characterized by its seed- 
less nature throughout most of the year. Frequently the fruit 
of the Kao Phuang and other seedy pummelos grown about 
Bangkok are seedless. This is generally true of the fruits that 
have developed from adventitious flowers, those flowers that 
have come latest and have been produced out of the usual flow- 
ering season. 

The very restricted area of about 48 square kilometers, de- 
voted to the production of the Nakorn Chaisri seedless pummelo, 
would indicate that the soil or the meteorological conditions are 
of a special nature in this section. The topsoil is a loose clay 
loam and the subsoil is a heavy gray clay which seems to be 
characteristic for the entire delta region. 

432 The Philippine Journal of Science i92i 

Some growers contend that the quality of the Nakorn Chaisri 
pummelo is due to the method of curing and storing. While 
this undoubtedly is a factor in producing a better flavor and 
aroma in all varieties of pummelos, it cannot account for the 
superiority of the Kao Pan over the other varieties. 


The Siam seedless pummelo, sometimes called the Nakorn 
Chaisri or Kao Pan pummelo, appears to be a distinct variety 
developed in a restricted region of the delta area in Ban Mai, 
Sarm Prarm, Nakorn Chaisri, Siam. Through the main part 
of the year it produces seedless fruits, but during November the 
fruit is reported to incline to seediness. 

The temperature in this region ranges between 24.4 and 32.1° 
C, the coolest months being December and January, and the hot- 
test about April. The mean daily range is about 9.3° C. The 
total rainfall for the district varies between 1,154.7 and 1,690 
millimeters during the year. A distinct dry season is present, 
usually starting in November and continuing through December, 
January, February, March, and April. 

The water used for irrigation in the Ban Mai orchards is taken 
directly from Tachin River. This water is extremely salty dur- 
ing at least six months of the year. In June analyses showed 
the presence of 2 per cent of salt, which approaches the salt 
content of sea water, with approximately 2.7 per cent of sodium 

The water used for irrigation in the Bang Bakok fruit section 
near Bangkok is obtained from the Menam Chao Phaya. The 
fruit produced in this section is seedy and its quality is inferior 
to that produced in the Ban Mai region. Analyses of this water 
showed that in June it contained only 0.04 per cent of salt, which 
is normal for ordinary river water. The water is reported to 
be less salty throughout the year than that of Tachin River at 
Ban Mai. 

The surface soil in all of the orchard sections ill the delta re- 
gion is heavy, clayey, and entirely of an alluvial formation. The 
subsoil is a heavy, gray, brick clay, which is rather impervious 
to water. 

Analyses of the salt content of the soils in Ban Mai show that 
the surface soil, subsoil, and canal mud used for fertilization, 
are extremely salty in June, containing between 0.8 and 2.5 per 

19, 4 Reinking and Groff: Siamese Seedless Pummelo 433 

cent of salt. The soil in the Bang Bakok region showed only 
a normal sodium chloride content, between 0.18 and 0.33 per 

The presence throughout the greater part of the year of a large 
amount of salt in the irrigation water, and consequently in the 
soil, of the Nakorn Chaisri region at Ban Mai seems to have a 
direct relation to the quality of fruit produced, and may have 
some relation to its seedlessness. The fruit produced in the 
Bang Bakok district, where the water is less salty throughout 
the year, is inferior in quality and very seedy. 

The cultural methods used in the best pummelo sections of 
Siam are well adapted to the low, wet, mud flats and nipa 
swamps of the region. The raised-bed methods followed are 
very largely those employed by the Chinese in the successful 
plantations of the flood-swept delta regions of southern China. 

The seedless pummelo trees flower and fruit throughout the 
year, but in most abundance four times each year. The largest 
number of flowers appear in June, at the beginning of the rainy 
season; from this lot of blossoms is produced the largest crop, 
ripening in November. The fruit in this crop is reported to 
have seeds. During the rest of the year no seeds, or only small, 
abortive ones, are produced. The fruit is well suited to storage 
and shipping. Records of storage and shipments covering a 
period of from six to eight months have been made. The fruit 
will remain in excellent condition for two or three months. 

The Kao Pan, or Nakorn Chaisri, pummelo has a well-rounded, 
usually low-headed, shapely tree. In form, shape of leaf and of 
fruit, and size and shape of oil cells, it somewhat resembles 
an orange. It is, however, a true pummelo. The fruit is fleshy, 
very juicy, white, and has a delightfully sweet, mildly acid 
flavor. It is probably the best of the pummelo types. 

The Nakorn Chaisri seedless pummelo is extremely delicate 
and, consequently, apt to be severely attacked by canker and 
other diseases and pests. The tree needs constant attention and 
care, but with such care excellent results can be obtained. 

Salting and cultural practices seem to have a direct relation 
to the quality of the fruit produced and possibly some relation 
to the seedlessness. 

The relation of seedless fruit to pollination was not fully de- 
termined. The soil and meteorological conditions may be factors 
in the production of excellent-quality fruit. 

434 1"^^ Philippine Journal of Science 

Curing and storage undoubtedly are factors in the production 
of better flavor and aroma in all varieties of pummelos. 

With the selection of a proper site and with proper cultural 
methods, the Nakorn Chaisri seedless pummelo could undoubtedly 
be grown successfully in the Philippine Islands and in the United 
States. The swamp lands near nipa-palm sections afford an ex- 
cellent location for experimentation with this fruit in the Phil- 
ippines. Regions in the Everglades of Florida should be 
desirable for the production of the Kao Pan, or Nakorn Chaisri, 
seedless pummelo. 


Plate 1 

Fig. 1. Tachin River near Ban Mai, showing nipa, Nipa fruticans Wurmb, 
swamp and thatched hut of an orchardist. 

2. Canal leading from Tachin River to the inland orchards. The river 
• is at low tide. 

3. Irrigation and drainage canal in the Nakorn Chaisri orchard section. 

Plate 2 

Fig. 1. Grass and nipa swamp at Ban Mai, Nakorn Chaisri. The land 
in the foreground has been cleared and is ready for the prepa- 
ration of mud dikes and canals. 

2. A raised dike along a canal in the Ban Mai region. The land is 

being prepared for pummelo culture. The embankment is a 
meter or less higher than the surface of the interior beds. 

3. Jak trees, Artocarpus Integra (Raderm.) Merr. {A. integrifolia 

Linn.), planted along the edge of a dike. Pummelo trees are 
planted within, on the first bed. 

Plate 3 

Fig. 1. Digging ditches in the preparation of the beds for the citrus orchard. 
The soil is used in the construction of the raised beds between 
the ditches. 

2. Preparation of ditches in the process of forming raised beds for 

pummelo culture. The soil is removed by hand after being cut 
into blocks with a special spade. 

3. Prepared beds in a newly developed plot planted to sugar cane 

before being ready for the citrus. 

Plate 4 

Fig. 1. Newly prepared beds planted to bananas before being ready for 
the citrus. 

2. Automatic water gate, leading from the canal through the dike 

into the ditches in the orchard. The pipe is made of a palm-tree 
trunk. The water is at; low tide. 

3. An older cultivated area in the Nakorn Chaisri region, showing 

the irrigation and drainage canal and the orchard land on both 

Plate 5 

Fig. 1. One of the best-managed Nakorn Chaisri pummelo groves at Ban 

2. Cement steps leading from a main canal up to the orchard house. 

3. Bruguiera sexangula (Lour.) Pers. and pineapples planted on a 

dike along a canal. The roots of the trees help to keep the earth 
at the base of the dikes in place. 


436 The Philippine Journal of Science 1921 

Plate 6 

Fig. 1. A walk on top of a dike; the side toward the canal is thickly 
planted with trees. 

2. A much-widened and deepened ditch in an old Nakorn Chaisri 

pummelo orchard. 

3. Trenching as carried on by Chinese labor in a Nakorn Chaisri 


Plate 7 

Fig. 1. Network of trenches, showing the effect of washing by rain in a 
Nakorn Chaisri orchard. 

2. Logs and bamboo poles thrown across the ditches to serve as foot 


3. Betel palms, Areca catechu Linn., growing along the edge of citrus 


Plate 8 

Fig. 1. Betel palms, Areca catechu Linn., growing along a walk in the 
citrus orchard. 

2. Betel palms, Areca catechu Linn., interplanted with citrus trees. 

3. Tall betel palms, Areca catechu Linn., scattered throughout a citrus 

plantation, providing a semishade for the pummelo trees. 

Plate 9 

Fig. 1. A much- widened ditch along the edges of which rice may be 
planted during the rainy season. 

2. Dead leaves, straw, and removed wood allowed to rot under the 

trees. Low-headed pumm-elo tree, branching from the surface of 
the ground. 

3. A pile of rice-paddy ash at the base of a pummelo tree in Nakorn 


Plate 10 

Fig. 1. A shapely, well-rounded, low-headed tree of a Nakorn Chaisri 
seedless pummelo. 

2. The lower branches of a Nakorn Chaisri seedless pummelo lying 

almost flat along the surface of the bed and extending over the 
trench of water. 

3. A low-headed, spreading tree of a Nakorn Chaisri seedless pummelo. 

Plate 11 

Fig. 1. Flowers and newly developed fruits formed on young stems, which 
shoot out quite upright from the branches of the previous year's 

2. A mass of old flowers and newly developed fruits produced from a 

single flower stalk. 

3. A flower of a Nakorn Chaisri seedless punmielo. 

19, 4 Reinking and Groff: Siamese Seedless Pummelo 437 

Plate 12 

Fig. 1. A Nakorn Chaisri seedless pummelo cut longitudinally and prepared 
ready for eating. 

2. A Nakorn Chaisri seedless pummelo cut open and each locul« 


3. A Nakorn Chaisri seedless pummelo cut longitudinally, showing 

thickness of skin, sections, and open center with the core ex- 
tending down one side. The fruit is not quite mature. 

Plate 13 

Fig. 1. Latitudinal section of the Nakorn Chaisri seedless pummelo shown 
in fig. 2. 
2. A Nakorn Chaisri seedless pummelo picked at Ban Mai about 
June 1, 1920; purchased at the orchard storehouse on June 8 
and shipped to the Philippine Islands. The picture was taken 
at Los Baiios on August 25, 1920. The fruit is still of excellent 
quality and flavor after long storage; diameter, 12.5 centimeters; 
skin, 8 to 10 millimeters thick; no seeds. 

Plate 14 

Fig. 1. Longitudinal section of the Nakorn Chaisri seedless pummelo 
shown in fig. 2. 
2. A Nakorn Chaisri seedless pummelo picked at Ban Mai between 
May 11 and 30, 1920; purchased at orchard storehouse on June 8 
and shipped to the Philippine Islands. The picture was taken at 
Los Banos on August 25. Fruit old and shriveled, but still 
of excellent quality and flavor; diameter, 11 centimeters; skin 
8 to 10 millimeters thick; no seeds. 

Plate 15 

Fig. 1. Black rot caused by a Diplodia on the fruit of a Nakorn Chaisri 
seedless pummelo. Rot produced in storage. 
2. Citrus canker caused by Pseudomonas citri Hasse on the fruit of 
a Nakorn Chaisri seedless pummelo. 

Plate 16 

Plat showing the cultural methods for the Siam Nakorn Chaisri seedless 
pummelo. Scale about 1 to 1,500. 

text figure 

Fig. 1. Map showing the location of the pummelo orchards at Ban Mai 
on Tachin River;, scale, about 1 to 530,000. 

Reinking and Groff: Siamese Seedless Pummelo.] [Philip. Joubn. Sci., 19, No. 4. 


Reinking and Groff: Siamese Seedless Pummelo.] [Philip. Journ. Sci., 19, No. 4. 


Reinking and Groff: Siamese Seedless Pummelo.] 

[Philip. Journ. Sci., 19, No. 4. 


Reinking and Groff: Siamese Seedless Pummelo.] [Philip. Journ. Sci., 19, No. 4. 


Reinking and Groff: Siamese Seedless Pitmmelo.] 

[Philip. Journ. Sci., 19, No. 4. 


Reinking and Groff: Siamese Seedless Pummelo.] [Philip. Journ. Sci., 19, No. 4. 


Reinking and Groff: Siamese Seedless Pummelo.] [Philip. Journ. Sci., 19, No. 4. 


Reinking and Groff: Siamese Seedless Pummelo.] [Philip. Journ. Sci., 19, No. 4. 


Reinking and Groff: Siamese Seedless Pummelo.] [Phtlip. Journ. Sci., 19, No. 4. 




Reinking and Groff: Siamese Seedless Pummelo.] [Philip. Journ. Sci., 19, No. 4. 

PLATE 10. 

Reinking and Groff: Siamese Seedless Pummelo.] [Philip. Joirn. Sci., 19, No. 4. 

PLATE 11. 

Reinking and Groff; Siamese Seedless Pummelo.] [Phiup. Journ. Sci., 19, No. 4. 

PLATE 12. 

Reinking and Groff: Siamese Seedless Pummelo.] [Philip. Journ. Sci., 19, No. 4. 

PLATE 13. 

Reinking and Groff: Siamese Seedless Pummelo.] [Philip. Journ. Sci., 19, No. 4. 

PLATE 14. 

Reinking and Groff: Siamese Seedless Pummelo.] [Philip. Journ. Sci., 19, No. 4. 

PLATE 15. 



L.*«f44.*-»•«t«^«^4.44^ [7^^.! ^^•♦'S^.i: ^ ^ -^ * * S •||jVj i: ^o + + 4 + 4^4 4 4 4 f ^M-^ 

i^rss!:^^:-zfzzr .^.:.jii: - >;^. : -T : ^--^=r ^ ■ ■ ^-. . '^=^ 1 - — ^ ^ '* "* L'^'y^ ^^*,_±_\ 4 4- 4 4 4 "^ * J + ^ 4. 4 4 4 4 y t r r t t ^ 4t^| 

prT^TTYf'^^'^-^-^'>^^^^y§'^ 4'i''rit^'~^'^"' ^ . ^ . . c§ . I 5 !•*• * Z' ^ ^ 4- 4> r r r . r 1 ^ ,M^A(^ 

li,.j,jujujc^jr_T^_jLJL^-^^ ^ ^ ^i^f'-^i^iul: ^ : w ; : ; : : : ^ " * r * -^n^ ^n- ^0— ^n^^- 1 

'fr^T-T-r^^r^^"^- f :fnr^ -> -^r"^^ k^,. : ^-4^ : ;^N ......;,., 4 -^ ^L 44 44,, J. ^ t^^ 

^ '^^<^444 4tTTTYj; 

^ ^ ^ * ^ ^ r ^ T ^ ^ f V -r r v^ *^ >»^ / ^ •* .» ^, /.. •♦ ■ 
Y Y t T T ^ r ^.A. '^ ^ ** -'^ ^ r ■»» <» >^ /< >M >'y ** ^ ■* '*j ; 

■■.A • <•■' •'v« ^ ^ ^ 

■■'■■' r ^^ ":: ' -r r ? 1^ r y 

in~~7~7~~i --"Xi 9 O Q ^5 If ir f Y r ^ ■,.>, ;.• ::• 'V r " t) 

4 4 4 4^4 4rf» 
4 4 4 4 -^-^o^ ^ .V 

4- + 

4 -^ 

4 4 

4 «4 

4 <4 

4 '4' 

4 4 

4 4 

4 4 

4 4 

4 4 

4 4 

^ 4 c4> .;. 4- ^ 4 4 44 

^ V f; ^- a -i i|- «*>„ ^ 4^ O ^ '*] 

<;; T r r^ 

^^^i^v.^pr^-'-^'T'^^^T^ ^X 4 ^^i^r-n 

4 * 4 4 o^ 4, ^ * 4 4 - -/ r J' -' T T--r-r\ Jj 

(?"■ ^ 4 4 4 •♦• 4 4' 4 4 '/-/'/': ^ ' ' ^ M,^ ' 4 

f^z +;~^ * + 4. ,4; 4 ^ -t* -f -^ 4 t T r T Tll^ ^o^. 

V"^. .'*',■ ^ *' . ^ ^' ^ J^ ^ 4 4 ■» 4 ^ T _ 7 T T TJ j ^■ 

~ ' ~ ' .i - . : — — .- y-siT Y 

f4 * 4 4.4^4'44, 4*.^447 r >- r OT 
1^ » 4 ,4 _. •*'__'*' 4^^ 444. 4.4 4T T_T rr] T 


rrr-rr-T--.. >. v v '/ '/ '/ '.' '/ '; '/ '-; '/ '/ ''\ -> ^ ^I;.'; 

T r f r 

T rTTrrrt-r 

r yr^rrrt^ 'i T Y r T r 

T Y r r T^ , ^ ^ 

r Y _T Y Y _r r Tj 

T~~Y Y rT~~^ ^~Y ^ ■, ^. 

r T Y r T 
T T r Y 

~7 r Y r~ Y r T Y Y r" 



~T f Y Y r ^ r' Y "^Y T~ r Y r" r r f "^1 ' 41 ?| 

Y Y Y Y Y Y Y T _ Y_ Y Y Y Y Y j | T 1 

"^ ■-- ^( 




C ^OD t_ 

c _c "S 

•8 '^ S IS gl 8 

fe o E .^ 2 '^ i5 

ooooooooodoo ooooo 

J- ■ vv r .-.-- V .^ - V ^^> ._^ .4; '.,-ii; 'V 

r" .^;. 


— rtf, 


liiii ill 


By Hans Gebien 
Hamburg^, Germany 


The great activity of Prof. C. F. Baker, as a collector on many 
of the islands comprising the Philippine Archipelago, has resulted 
in bringing so much new material in the Tenebrionidse to the 
attention of science that it is believed worth while to make this 
material the subject of a special paper. I regret that I cannot, 
with the Philippine material collected up to the present, comply 
with Professor Baker's request to make a synopsis of the Phil- 
ippine Tenebrionidse. The fact that this zealous collector h^s 
succeeded in the short space of three years ^ in bringing together 
so large a number of new species demonstrates that it would be 
premature to undertake this task at present; very much more 
new material will surely be found when the mountainous and in- 
accessible parts of the Islands, especially those that are infre- 
quently visited, shall have been explored. It is especially neces- 
sary, for zoogeographical reasons, that the more remote islands 
be explored — that is, such as lie near other faunal regions — in 
order to determine to what extent the forms from those regions 
intergrade with those of our own. Furthermore, systematic 
work on Indo-Malayan Tenebrionidse is still very obscure and 
cannot be attempted with material from a restricted faunal area. 

As we now have more than one hundred fifty species of Tene- 
brionidse a fairly clear idea can be formed of the Philippine 
tenebrionid fauna. Evidently we are here dealing with a pure 
Indo-Malayan fauna. Naturally, there is no lack of genera 
that are restricted to the Philippines. Where such is the case 
the nearest relationship must be sought, almost without ex- 
ception, on the neighboring islands. To these endemic genera 
belong Oedemutes, Pseudostrongylium, Aptereucyrtus, Pseuda- 
bax, and Lophocnemis. Only a few genera stand entirely 
isolated, no related genera being found in other faunal regions ; 

* For Part I, see Philip. Journ. Sci. § D 8 (1913) 373-433. 

* This paper was completed early in 1916. 


440 ^^^ Philippine Journal of Science 1921 

for instance, Allopezios and Bolitrum, These, of course, do not 
lend themselves to zoogeographical study. By far the greater 
number of genera is represented only on the Sunda Islands, or 
possibly in the interior of India, Formosa, and Ceylon. These 
are Bolitoxenus, Leiochrodes, Setenis, Encyalesthus, Catapiestus 
Artactes, Scotaeus, Pseudeumolpus, Platycrepis, Eucyrtus, Sima- 
lura, Hemicera, Psydus, Camarimena, Pseudonautes, Gauromaia 
Dietysus, and Aediotorix. There remains a considerable num 
ber of genera the species of which extend over a wider region 
These genera, to which belong Mesomorphiis, Bradymems 
Byrsax, Ceropria, Cossyphus, and Lyprops, extend partially 
from Africa to Australia. Scleron finds here its most easterly 
extension, Cnemodasus its most westerly. 

Only two genera fall entirely beyond these limits. As to the 
first one. Ethos, I doubt that it occurs in the Philippines ; despite 
exhausted search in the Islands it has not been found again since 
Eschscholtz'3 time. It is possible that the habitat was incor- 
rectly reported and that the animals in question were found in 
India, where it occurs. Such an oversight can easily be under- 
stood when we consider that Kotzebue in his journey around the 
world touched many different regions. The second genus, 
Leptoscapha, I have discussed elsewhere. 

The foregoing remarks lead to the conclusion that the fauna 
of the Philippines must belong to the Indo-Malayan region. 

I am greatly indebted to Professor Baker, whose untiring 
activity as a collector made possible the preparation of this 
paper, and who most generously presented to me single speci- 
mens for my own collection. To express to him here my hearty 
thanks is an agreeable duty. Further, several new species were 
found in the museums of Stettin, Dresden, and Hamburg, as well 
as in my own collection. I am also greatly indebted to Mr. P. 
Timm, member of the Chamber of Audits in Zoppot, who not 
only presented me with several species from his fine collection 
but also, by means of excellent photographs taken by him for me 
at the expenditure of much time, enhanced the value of this 
work not a little.^ 


Mesomorphus maquilingius sp. nov. 

Klein, sehr gewolbt, dunkel, matt schwarzbraun, Fiihler und 
Beine gelblich braun. Der Clypeus ist halbkreisformig ausge- 

3 The introduction was translated from the German by the Bureau of 
Science. — Editors. 

19, 4 Gebien: Philippine Tenebrionidse, II 441 

schnitten, die Seitenlappen halbkreisformig, der Vorderrand 
des Kopf es also wie bei M. villiger; die Oberlippe mit feinem Aus- 
schnitt am Vorderrand. Der Kopf ist flach, der Quereindruck 
sehr schwach begrenzt; am Innenrand der Augen befindet sich 
eine schmale, aber kraftig erhohte Langsfalte, viel starker als 
bei villiger, die durch darauf gestellte Wimperborsten noch deut- 
licher wird, und namentlich bei der Ansicht von der Seite 
auflfallig ist. Die Skulptur besteht aus groben, flachen Punkten, 
deren Zwischenraume ein feines, ziemlich regelmassiges Netz- 
werk von glanzenden Erhabenheiten bildet; jeder Punkt hat im 
Zentrum eine kurze, fast aufrechte Borste. Die Wangen sind 
vor den Augen viel breiter als bei villiger, ihre Ecken abge- 
rundet; die Bildung erinnert mehr an Gonocephalum als an 
Mesomorphus. Die Fuhler sind kurzer als bei villiger, ihre vor- 
letzten Glieder starker quer und das letzte ist nicht verlangert, 
sondern nur so lang wie breit. 

Der Halsschild ist der Quere nach stark gewolbt, seine 
Seitenrander sind gleichmassig, ziemlich stark verflacht, die 
Vorderecken sind verrundet rechtwinklig, die grosste Breite 
liegt hinter der Mitte, die iseitliche Rundung ist sehr stark, die 
Hinterecken sind scharf rechtwinklig, jederseits des Mittellap- 
pens findet sich eine breite Ausbuchtung, so dass die Winkel 
nach hinten gerichtet erscheinen ; der basale Mittellappen reicht 
weiter zurtick als die Ecken. Die Punktierung und Beborstung 
ist wie auf dem Kopf, die Borsten sind halb aufrecht, nach 
hinten gekriimmt und viel kurzer als bei M. villiger, so dass eine 
vordere Borste den Grund der hinteren nicht bedeckt. Das 
Schildchen ist ziemlich blank und fein punktiert. 

Die Flugeldecken sind kurz und sehr stark der Quere nach 
gewolbt, ihr Seitenrand von oben nicht sichtbar ; die Beborstung 
ist unregelmassig zweizeilig, vor der Spitze aber in den etwas 
schmaleren zweiten, vierten, und sechsten Zwischenraum ein- 
zeilig, bei M. villiger dagegen ebenfalls zweizeilig, wenn auch 

Die Unterseite ist nicht heller als die Oberseite, aber viel f einer, 
anliegend beborstet. Die Vorderbrust ist vom und an den 
Pleuren mit scharfen, glanzenden, runden Kornern versehen 
(bei M. villiger punktiert), der Prosternalforstsatz ist hinten 
niedergedriickt ; im ubrigen ist die Unterseite ganz ahnlich; die 
Vorderschienen sind schmaler als bei dieser Art. 

Luzon, Laguna, Mount Maquiling, 1 Exemplar. 

Lange, 5.6 Millimeter ; Breite, 2.7. 

442 The Philippine Journal of Science 1921 

Ich habe diese Art mit M. villiger verglichen, weil diese in 
alien Sammlungen vertreten und sehr gemein ist, sich auch auf 
den Philippinen findet; naher aber steht ihr eine neue Art von 
Blrmah aus meiner Sammlung, die einen schmalen Canthus hat 
und deren Zwischenraume hinten deutlich punktiert sind, auch 
bei dieser ist die Vorderbrust deutlich gekornt. Sehr ahnlich 
ist auch M. picescens aus Abessinien aber schmaler, matt 
schwarz, die Streifen sind an der Spitze grob punktiert, und das 
Mentum ist nicht gehockert. 

Mesomorphus villiger Blanch. 

Weiteres Material liegt mir vor von Palawan, Puerto Prin- 
cesa; (2^57)^ auch von Luzon, Mount Banahao, Mount Maqui- 


Cnemodasus rectangulus Geb.^ 

Davon liegen mir zwei Exemplare von Los Banos vor (liS7), 
ferner ein Exemplar von Manila (Hallier leg. XI.-XIL 1903) im 
Mus. Hamburg. 

Gonocephalum bilineatum Walk. 

Luzon, Benguet, Baguio (4990). Palawan, Puerto Princesa 
(4051). Auch von Mindanao (Peters leg.). 


Genus BRADYMERUS Perroud 

Obgleich die Arten dieser Gattung im allgemeinen in der 
Skulptur und Kopfbildung sehr gute Merkmale haben, ist ihre 
Deutung durchaus nicht immer leicht, da die Beschreibungen 
von Fairmaire und Schaufuss meist ganz unzureichend sind. 
Meine Auffassung von B. crenulicollis ist, durch die mangelhafte 
Beschreibung verursacht, eine falsche. B. crenulicollis ist 
durch die spitzig vorragenden Wangen und die Skulptur sehr 
ausgezeichnet. Ich habe ihn noch einmal als B. denticeps 
beschrieben. Die von mir als B. crenulicollis ^ aufgefasste Art 
steht dem B. elongatus Perty (=B. javanus Fairm.) nahe und 
wird hier neu beschrieben als : 

* Die hier angegebenen Nummem sind diejenigen unter denen mir Herr 
Baker die Arten mitteilte. 

• In meiner Arbeit im Philip. Joum. Sci. § D 8 (1913) 373 gehort Zeile 
25-31 ("Die verkurzten Epipleuren-unterscheidet") zur Gattungsbeschrei- 
bung von Cnemodastis. Ich hatte seinerzeit keine Korrektur gelesen. 

•Bull. Sarawak Mus. 2 (1914) 11. 
'Philip. Journ. Sci. § D 8 (1913) 379. 

19.4 Gebien: Philippine Tenehrionidm, II 443 

Bradymerus mcgregori sp. nov. 

Schmal, parallel, gewolbt, hellbraun, Fltigeldecken glanzend. 
Der Kopf ist flach, die Clypealsutur breit und flach eingedruckt, 
die Wangen sind ganz verrundet und nicht breiter als die Augen, 
die Augenfalten sind schmal, wenig deutlich, und gehen im 
Bogen urn das Auge herum. Die Punktierung ist dicht und 
deutlich, die Zwischenraume der Punkte sind kurze, scharf 
erhabene Langskielchen ; der Nacken ist fein gekornt, der Cly- 
peus vorn ganz gerade abgestutzt. Die Fiihler sind schlank 
und haben eine starke, 6-gliedrige Keule, die vorletzten Glieder 
sind reichlich anderthalbmal so breit wie lang, das letzte so breit 
wie lang. Die Mandibeln sind an der Spitze tief gefurcht, das 
Mentum scharf gekielt. 

Der Halsschild ist doppelt so breit wie lang, die Scheibe hoch 
gewolbt, aber die Mitte der Lange nach flach gedruckt. Die 
Seiten sind schwach gerundet, nach hinten wenig verengt, daher 
die Basis viel breiter als die Spitze, der Rand ist fast glatt, die 
Mitte des Vorderrandes ist breit lappenformig nach vorn gezo- 
gen, die Vorderwinkel ragen spitz und lang vor, die Seiten sind 
schmal und in der Mitte etwas grubig vertieft abgesetzt. Die 
Punktierung ist grob, ganz dicht gedrangt, doch bilden die 
Zwischenraume keine Korner, aber vorn in der Mitte feine 
Langserhabenheiten. Die Hinterwinkel sind scharf recht- 
winklig, die Basis ist stark, in der Mitte breiter gerandet. 

Der Seitenrand der Fltigeldecken ist von oben sichtbar, die 
altemierenden zwischenraume sind scharf kielformig erhaben, 
der achte lauft hinten bis in die Spitze, wo er sich mit dem 
ersten verbindet; der scharf e Kiel des ersten Interstitiums 
beginnt im letzten Drittel, die Kiele des dritten, funften, und 
siebenten beginnen hart an der Basis, sie sind nahezu glatt. 
Im vierten und sechsten Zwischenraum zeigen sich sehr 
schwache Spuren von Zwischenkielen, der Grund der Fltigel- 
decken zeigt mikroskopische Nabelpunkte. 

Die Unterseite, besonders das Abdomen, ist staubartig be- 
haart, die Vorderbrust und die Propleuren sind sehr grob 
punktiert, das Prosternum ist der Lange nach gewolbt, der 
Absturz aber zuletzt senkrecht; das Mesostemum ist breit 
U-formig ausgerandet, die Hinterbrust vorn und seitlich grob 
punktiert. Die Beine sind schlank, die Schienen dtinn, die 
hinteren zeigen in ihrer Endhalfte einen feinen Haarstreifen. 

Lange, 7.6 Millimeter; Breite, 3.1. 

Luzon, Benguet, Irisan River. Ein zweites Exemplar muss 
sich in der Sammlung des Bureau of Science befinden. Ferner, 

444 The Philippine Journal of Science 1921 

vier Exemplare aus dem Museum Stettin von den Philippinen 
{Semper leg.). 

Ich hielt diese Art zuerst nach der Beschreibung ftir B, ere- 
nulicollis Fairm. ; sie steht ihm aber wegen der schmalen Wangen 
sehr fern. Am Nachsten verwandt ist B. elongatus Perty {=B. 
javanus Fairm.), der aber einen krenulierten Halsschild- 
seitenrand hat, kraftig und scharf entwickelte Zwischenrippen, 
und dessen Hinterschienen einfach sind. 

Bradymems pertyi nom. nov. 

Bradymerus elongatus Geb. muss wegen B, elongatus Perty neu 
benannt werden ; ich nenne ihn B, pertyi nom. nov. 

Bradymerus ferruginipes Fairm. 

Luzon, Mount Banahao {WW), 2 Exemplare. 

Bradymerus altemicostis Geb. 

Luzon, Mount Maquiling {29 33), 

Bradymerus clathratus Schauf. {=aequecostatus Fairm.). 

PALAWAN, Puerto Princesa {i023) , Luzon, Tayabas, Ma- 
linao {5430), Mount Banahao. Mindanao, Dapitan. Weiteres 

Bradymerus eschscholtzi sp. nov. 

Schwarzblau, fast schwarz, die Decken blaugrun, Unterseite 
und Beine glanzend schwarzbraun ; lang gestreckt, fast parallel. 
Der Kopf ist lang und nahezu flach, doch ist die Clypealsutur 
fein und deutlich eingedriickt und ihre Ausmlindungsstelle am 
Seitenrand des Kopfes durch einen feinen Ausschnitt gekenn- 
zeichnet, der Vorderrand ist ganz gerade abgestutzt, nicht wie 
bei B. carinatus deutlich ausgebuchtet. Die Augenfurchen sind 
sehr tief und laufen hinten in den Nacken, nicht um das Auge 
herum. Die ersten Glieder der Fiihler sind rotbraun, die letzten 
funf bilden eine gut abgesetzte Keule. ' Das Mentum ist fein 
der Lange nach gekielt, die Mandibeln an der Spitze nur sehr 
undeutlich gefurcht. 

Der Halsschild ist reichlich anderthalbmal so breit wie lang, 
die Seiten sind schwach gerundet und fast glatt, nur ganz un- 
deutlich uneben, der Rand ist sehr schmal abgesetzt, die Vor- 
derwinkel ragen lang und spitz vor, die Hinterecken sind scharf 
rechtwinklig, die Seiten sind nach hinten nur ganz schwach und 
fast geradlinig verengt, die Mitte des Vorderrandes ragt nicht 
vor, die Scheibe ist gleichmassig stark gewolbt und hat keinen 
Langseindruck, die Punktierung ist ziemlich grob und nur in 
den Vorderwinkeln gedrangt, zwischen den Punkten befinden 

19,4 Gebien: Philippine Tenebrionidse, II 445 

sich zahlreiche klejne, flache Kornchen, die ein mikroskopisches 
Punktchen tragen; die Punktierung ist etwas grober und weit- 
laufiger als die des Kopfes, die Basis ist nicht deutlich geran- 
det, nur in der Mitte etwas verflacht abgesetzt. 

Der Seitenrand der Flugeldecken ist in der hinteren Halfte 
von oben nicht sichtbar, sie sind stark gewolbt; die Punkte der 
Streifen sind kraftig, gleichmassig und durch eine feine, ein- 
gedruckte Linie miteinander verbunden. Von den Streifen ist 
der erste hinten nur stark gewolbt, der dritte in der Endhalfte, 
der funfte vollstandig gekielt, der sechste ebenfalls aber hinten 
abgeklirzt, der siebente ganz gekielt, der achte in der End- 
halfte; dieser ist aber nicht bis zur Spitze fortgesetzt wie 
bei sehr vielen Arten, sondern es findet sich dort nur eine quere 
Wolbung; der dritte und vierte tragen in der vorderen Halfte 
sehr flache, langliche Kornchen. 

Die Unterseite zeigt keine Spur von Harchen, die Vorder- 
brust ist in der Mitte und auf den Pleuren grob punktiert, das 
Prosternum ist breit gefurcht, der Fortsatz etwas niederge- 
druckt, die Hinterschienen sind in der Mitte der ganzen Lange 
nach glatt und jederseits mit einem mikroskopischen Haar- 
streifen versehen. 

Lange, 9.2 Millimeter; Breite, 4. 

Luzon, Manila (Eschscholtz), 2 Exemplare, in meiner Samm- 
lung und im Berliner Museum. 

Ich hielt diese Art ursprlinglich fiir eine Varietat des ver- 
anderlichen B. carinattcs Fairm. Eine genauere Priifung zeigt 
jedoch, dass wir es sicher mit einer guten Art zu tun haben, 
deren Hauptmerkmale die tiefen, in den Nacken gehenden 
Augenfurchen, der geradlinig abgestutzte Clypeus, der feine 
Kiel des etwas anders gestalteten Mentums, und etwas andere 
Skulptur sind. 

Bradymerus carinatus Fairm. 
Luzon, Mount Banahao i5J^29) : Tayabas, Malinao. 

Tabelle der Bradymerus-Arten der Philippinen. 
Die vorstehenden Bemerkungen liber Bradymerus und die 
Wiederauffindung von B, ferruginipes Fairm. lassen eine neue 
Bestimmungstabelle vOn den philippinischen Arten wiinschens- 
wert erscheinen. Die Arten lassen sich folgendermassen 
auseinanderhalten) : 

1. Blaue Oder metallische Arten, Fuhlerkeule 5-gliedrig 2. 

Oberseite schwarz oder braun, Fuhlerkeule 6-gliedrig 5. 

446 The Philippine Journal of Science 1921 

2. Halsschild weitlaufig punktiert, sein Vorderrand gerade abgeschnitten, 

Fliigeldecken ungekielt, die Punkte der Streifen sehr grob, seitlich 

griibchenartig B. caeruleipennls Geb. 

Halsschild grob und dicht punktiert, Vorderwinlcel vorragend, Fliigel- 
decken gekielt, die Punkte fein ^ .3. 

3. Oberseite leuchtend blau, Prosternum hinten ganz niedergedriickt. 

B. violaceus Pasc. 
Oberseite griinlich oder bronzefarben, Prosternum nicht niederge- 
driickt 4. 

4. Augenfalten fein und um das Auge herumgehend, Clypeus deutlich aus- 

gerandet, Vorderkorper griinlich, Fliigeldecken bronzefarben. 

B. carinatus Fairm. 
Augenfalten sehr grob und hinten sich von den Augen entf ernend, Vor- 
derrand des Kopfes gerade abgestutzt, Korper vorn schwarzblau, Flii- 
geldecken blaugriin B. eschscholtzi sp. nov. 

5. Die Wangen sind seitlich scharfwinklig vorgezogen; sehr gedrungene 

Art mit breit abgesetzten Seitenrandern, die Beine ganz rot, die 
Schienen aussen fein gekielt, Oberseite matt schwarz. 

B. ferruginipes Fairm. 

Die Wangen verrundet und nicht breiter als die Augen; schlanke Arten 

mit schmal abgesetzten Halsschildrandern, die Schienen ungekielt.. 6. 

6. Langgestreckte Arten, Stirn fein langsstrigos 7. 

Kiirzere, gedrungenere, normal gestaltete Arten, Stirn kornig punk- 
tiert 9. 

7. Halsschild in der Mitte mit Eindruck, Vorderwinkel ziemlich spitz vor- 

ragend, alle Streifen vorn gekielt, Fiihlerkeule schwarz. 

B. impressicolUs Geb. 

Halsschild gleichmassig flach gewolbt, Vorderwinkel schwach, gerundet 

vorragend, die abwechselnden Streifen erhabener, Fiihler ganz rot.... 8. 

8. Grund der dunkelbraunen Flugeldecken spiegelglatt, die Rippen auf der 

Scheibe fast glatt B. alternicostis Geb. 

Grund der schwarzen Fliigeldecken sehr fein lederrunzlig, die Rippen 
fein gekornt {elongatus Geb.) B. pertyi nom. nov. 

9. Die alternierenden Streifen stark erhaben; rotbraune Art. 

B. mcgregori sp. nov. 

Alle Zwischenraume bis auf die inneren gleichmassig stark erhaben; 

oben fast schwarze Art B. aequecostatus Fairm. 

Bolitoxenus ditylus sp. nov. 

Kurz zilindrisch, oben etwas abgeflacht, matt schwarz. Der 
Kopf ist flach ausgehohlt, die Clypealsutur sehr fein und kaum 
eingeschnitten ; Horner finden sich nicht auf dem Kopf, dagegen 
verschiedene Erhebungen, die f olgendermassen angeordnet sind : 
auf dem geradlinig abgestutzten Clypeus finden sich beim Mann- 
chen zwei kurze konische Tuberkeln, beim Weibchen zwei quere, 
auf der oberen Kante krenulierte Kiele, die durch einen schmalen 
Zwischenraum getrennt sind, also auch als ein unterbrochener 
Kiel angesehen werden konnen, die Wangen treten spitz zahn- 
formig nach aussen, die Seiten des Kopfes davor haben einige 

19, 4 Gebien: Philippine Tenebrionidse, II 447 

stumpfere Zahne und sind etwas buckelig. Am Hinterkopf be- 
findet sich eine quere Reihe von vier spitzen Tuberkeln, von 
denen die beiden ausseren am inneren Hinterrand der Augen 
stehen. Beim Weibchen finden sich vor den beiden mittleren auf 
der Stirn noeh zwei kleinere. Die Punktierung der Stirn ist 
fein und sehr dicht, die des Hinterkopfes hinter den Kornern 
grob und sehr dicht. Die Wurzel der Fuhler und die Taster 
sind braunrot, die Fuhler sind scheinbar 10-gliedrig da das kleine 
Endglied in dem breiten Spitzenausschnitt des zehnten Gliedes 
eingefiigt ist und mit ihm zusammen etv\ra halbkreisformigen 
Umriss hat, die Glieder sind vom funften an mehr oder w^eniger 
nach innen erweitert. Das Endglied der Maxillarpalpen ist 
zilindrisch, das Mentum flach gewolbt ; der Aussenrand des Unter- 
kopfes neben den Augen its hoch gekielt und durch eine tiefe 
Rinne von den Augen getrennt. 

Der Halsschild ist ungefahr doppelt so breit v^ie lang, hoch 
gewolbt, der Rand sehr breit verflacht und auch hinten scharf- 
kantig. In der Mitte des Vorderrandes erheben sich zwei starke, 
kurze, an der Spitze verrundete Horner deren Zwischenraum 
beim Mannchen ein parallelseitiger Schlitz ist, beim Weibchen 
sind es zwei uber halbkreisformige Beulen, beide Bildungen 
einander sehr ahnlich, nur beim Weibchen kiirzer als beim 
Mannchen; die Homer stehen beim Mannchen am Vorderrand, 
beim Weibchen etwas entfernt davon. Der Rand ist hinter der 
Mitte am breitesten, mit neun bis zehn starken Sagezahnen ver- 
sehen, die vorn etwas kleiner werden ; die spitzen Vorderecken 
ragen weit vor, die Hinterecken sind scharf stumpfwinklig, hin- 
ter jedem vor deren Hornchen findet sich eine fast kreisformig 
angeordnete Gruppe von runden, scharf en Komchen und jeder- 
seits am Absturz noch einige einzeln stehende Kornchen; der 
Grund des Pronotums zeigt nur beim Mannchen auf der Mitte 
zeimlich f eine Punkte. 

Die Flugeldecken sind hoch gewolbt, oben etwas abgeflacht, 
sie haben Reihen regelmassiger, grober Punkte, welche meist 
durch die Tuberkeln aus ihrer Richtung gedrangt werden ; der 
Seitenrand ist kraftig, etwas ungleichmassig gezahnt, die Zahne 
aussen kurz verrundet, die Zahnreihe durch eine feine vertiefte 
Linie abgesetzt. AUe Zwischenraume sind gekomt oder gehock- 
ert, die beiden haben eine ziemlich regelmassige Reihe feiner 
spitzer Komer, der dritte und der f unfte langliche scharfe Tu- 
berkeln, der vierte, siebente, und so weiter, sind Ahnlich wie die 
beiden ersten skulptiert, der siebente hat grobere Korner; im 

448 The Philippine Journal of Science 1921 

ubrigen sind die Interstitien unpunktiert. Die Zahne des Sei- 
tenrandes laufen, immer feiner werdend, bis zur Spitze. 

Das Prostemum ist wagerecht, fast kielf ormig, und f allt hinten 
fast senkrecht ab, nach vorn noch steiler als hinten; das Meso- 
sternum hat in der Mitte und das Metasternum dicht dahinter 
vorn eine spitze Tuberkel. Die ganze Unterseite ist ziemlich 
grob und wenig dicht punktiert. Alle Schenkel haben unten 
zwei scharfe Leisten, die Schienen sind aussen scharf gekielt 
und werden gegen die Spitze dunner, die Fusse sind sehr zart. 

Lange: Mannchen, 7.8 Millimeter; Weibchen, 9.8. Breite: 
Mannchen, 4 Millimeter; Weibchen, 5. 

Luzon, Laguna, Mount Maquiling, ein Parchen. 

Ich glaube die beiden Tiere richtig als Angehorige einer Art 
zu bezeichnen. Die Geschlechtsunterschiede liegen in der Kopf- 
und Halsschildbildung. 

Die beiden andern Arten von den Philippinen, Bolitoxenus 
(Atasthalus) serratiis Geb. und B, timmi sp. nov., haben zwei 
lange Horner auf dem Halsschild, grobe Hocker und spitze 
Tuberkeln auf den Decken und ein ganz anderes Prosternum. 
Aehnlicher ist J?, spectabilis Geb. von Borneo, aber viel breiter, 
beim Mannchen mit zwei spitzen hornchenartigen Tuberkeln auf 
dem Halsschild ausgezeichnet, und hat anderes Prosternum und 
andere Kopfbildung. 

Bolitoxenus timmi sp. nov. Tafel 1, Fig. 1 und la. 

Gross, gewolbt, parallelseitig, ganz mit den braunen Ueber- 
resten des Wirtspilzes bedeckt, Fuhler und Beine braun. 

Der Kopf ist breit, flach ausgehohlt, der Clypeus aufgebogen 
und mit zwei stumpf en Winkeln am Vorderrande versehen ; die 
Wangen treten blattformig, zackig nach aussen, zwischen ihnen 
und den Winkeln des Clypeus finden sich ausser einem grossen 
stumpfen Winkel einige f eine Zahnchen. Bei schwacher entwick- 
elten Mannchen erscheint der Vorderrand einfach sehr schmal 
aufgebogen und fein krenuliert, auch sind die Wangen nicht so 
scharf ausgezogen. Auf der Stim finden sich keine Tuberkeln, 
nur am Innenrand der Augen ein spitziges Hockerchen in 
beiden Geschlechtern. Die Fuhler sind scheinbar 10-gliedrig, 
schlank, zu einer kraftigen Keule verdickt; das zehnte Glied 
ist auf der Spitze tief ausgeschnitten und nimmt das elfte, sehr 
kleine, fast kugelige Glied in seinem Ausschnitt auf. Die Taster 
sind gelbrot, das Endglied der Labialpalpen ist nackt, nicht wie 
bei B, serratus beim Mannchen mit einem langen Haarpinsel 
versehen; das Kinn ist flach gewolbt. 

19, 4 Gebien: Philippine Tenebrionidse, II 449 

Der Halsschild hat beim Mannchen zwei lange, wagerechte, 
nach vom gerichtete, mit der Spitze gegeneinander gewendete 
Homer, die bei wohlentwickelten Exemplaren an der Spitze ein 
Haarbiiehsel tragen und in der Endhalfte innen ebenfalls gold- 
braun behaart sind. Das am schwachsten entwickelten Mann- 
chen ist asymmetrisch und tragt nur auf dem linken Horn viel 
schwachere Behaarung. Beim Weibchen finden sich zwei (iber 
halbkreisformige, kraftige, etwas divergierende, nach vorn ge- 
richtete Hocker, die wie die Horner stark gekornt sind. Die 
Scheibe ist ebenfalls stark, etwas unregelmassig gekornt, und nur 
die aussersten Seiten sind frei, diese sind sehr breit verflacht 
abgesetzt, der Seitenrand selbst mit etwa 9 bis 10 kraftigen, 
rechtwinkligen, also nicht fingerformigen Zahnen versehen, die 
Hinterecken sind scharf stumpfwinklig. 

Die Fliigeldecken haben einen sehr fein gekerbten Basalrand, 
sie sind auf der Scheibe etwas flach bis zum dritten Zwischen- 
raum, ihre Skulptur ist durch die anhaftenden Teile des Wirts- 
pilzes ganz bedeckt und nur nach griindlicher Reinigung 
sichtbar, sie lassen sich aber am trockenen Kafer leicht mit 
einer spitzen Nadel abkratzen. Dann erkennt man, dass ziemlich 
regelmassige Reihen grober, runder, tief eingedriickter Punkte 
vorhanden sind, deren glatte Zwischenraume sammtlich gekornt 
sind, und zwar mit je einer Reihe von verschieden grossen Kor- 
nern oder Hockem; der Nahtstreifen mit sehr f einen, wenig 
engen, runden Kornern, der zweite mit etwas groberen, spitzen, 
konischen, der dritte mit etwa 4 etwas langlichen bis zum Ab- 
sturz und einigen kleineren dahinter; von diesen ist der Hocker 
an der Basis stark langlich und besteht aus 2 bis 4 eng gestellten 
Kornern. Vom vierten ab sind die alternierenden Zwischen- 
raume mit groberen und feineren sehr weitlaufig gestellten 
Kornern besetzt. Der Seitenrand ist einfach kraftig gesagt, die 
Sagezahne etwa rechtwinklig, nicht fingerartig. 

Das Prostemum ist ganz wagerecht, der Lange nach scharf 
gekielt, ebenso die Mittelbrust; beide fallen steil, aber nicht 
senkrecht ab, so dass ein kleiner Winkel zwischen den Kielcn 
entsteht, wodurch sich eine Neigung des Vorderkorpers gegen 
den Hinterkorper ermoglichen lasst, die Pleuren sind scharf 
und fein gekornt, das Abdomen ist grob punktiert. Die Schen- 
kel unten mit doppelten Kielen, die Schienen sind gerade, a'ussen 
mit 3 scharfen Kielen versehen, die Tarsen sind kurz. 

Lange, 7.8 bis 9.2 Millimeter (ohne Horner) ; Breite, 4.3 
bis 5. 

450 ^^^ Philippine Journal of Science 1921 

Sieben Mannchen, elf Weibchen von Manila aus der Sammlung 
von Rechnungsrat Tinun in Zoppot, Klimowitz, Zoppot, Strecker, 
Zoppot (von Herrn Peters gesammelt). 

Ich benenne diese Art zu Ehren des tuchtigen Entomologen 
und Photographen, der diese Arbeit durch seine Sammlung und 
die Herstellung der schonen Bilder reich unterstiitzte. 

Nur mit B, serratus verwandt, aber grosser, der Clypeus 
ohne Horner, die Reihen der Punkte auf den Decken nicht durch 
die Hocker unterbrochen, die Seitenrander einfach und eng ge- 
zahnt, die Hocker bis auf die des dritten Zwischenraumes klein. 

Byrsax satanas Geb. Tafel 1, Fig. 2 und 3. 
Neues Material liegt nicht vor. 


Platydema marseuli Lew. 

Luzon, Laguna, Mount Maquiling, 1 Mannchen. In Japan 
haufig, auch von Tonkin, und Borneo bekannt. 

Platydema malaccum Mars. 

Palawan, Puerto Princesa, 2 Weibchen; auch von Borneo, 
Java, Sumatra. Das Mannchen hat charakteristisch gekrummte 
Mittel- und Hinterschienen wie P. umbratum Mars. Ich glaube, 
dass P. annamitum sich von unserer Art nicht trennen lasst. 
Sicher ist auch P. laticorne Fairm. nichts anderes. 

Ceropria induta Wied. 

Palawan, Puerto Princesa U050). Mindanao, Butuan 
(i0i6). LEYTE, Tacloban. Weiteres Material. 

Ceropria subocellata D. & Br. 
Mindanao, Butuan und Iligan H0i5, h0h7, UOUS, WJf9). 


Leiochrodes (subg. Leiochrota) pMlippinensis sp. nov. 

Von fast kreisf ormigem Umriss, nur nach vorn etwas verlan- 
gert, glanzend rotbraun, durchscheinend, auf der Scheibe etwas 
dunkler, die Fuhler schwarz mit rotbraunen Basalgliedern und 
gelbem Endglied. 

Der Kopf ist flach und in eine parallelseitige Schnauze aus- 
gezogen, vollkommen unpunktiert, die Fuhler sind sehr lang und 
uberragen die Basis des Halsschildes weit, Glieder 3 und 4 sind 
nicht erweitert, die folgenden Glieder sind gleich, etwas langer 
als breit, sammtlich dick und ziemlich lang gestielt. 

19.4 Gebien: Philippine Tenebrionidse, II 451 

Der Halsschild ist stark nach vorn verengt, seitlich fast gerad- 
linig, der Vorderrand von oben gesehen kaum ausgeschnitten, 
die Vorderecken kurz verrundet stumpfwinklig, die Hinterecken 
sehr spitz und etwas nach hinten gezogen, die Oberflache ist 
ganz glatt, ohne Spur yon Punkten. 

Die Fliigeldecken sind kreisformig, an der Basis so breit wie 
der Halsschild, setzen aber seinen Umriss nicht fort, sie sind 
spiegelglatt, der Seitenrand ist an der Schulter schwach ver- 
breitert. Das Prosternum ist breit, wagerecht und glatt, die 
iibrige Unterseite unpunktiert, die Tarsenglieder sind kurz ge- 
lappt, an den Hintertarsen ist der Lappen des vorletzten Gliedes 
kaum halb so lang wie das letzte Glied. 

Lange, 4.4 Millimeter. 

Luzon, Laguna, Los Banos, 1 Exemplar. 

Die langen Fuhler weisen dieser Art einen Platz in der Unter- 
gattung Leiochrota an. Sie steht der L. uniformis Westw. sehr 
nahe, aber die Fuhler sind noch langer, die Glieder stark gestielt, 
und die ganze Oberseite ist glatt, anstatt punktiert. 

Uloma orientalis Cast. 

Palawan,'' Puerto Princesa. Luzon, Tayabas, Malinao; 2 

Uloma contracta Fairm. 

Luzon, Laguna, Mount Maquiling. 

Alphitobius diaperinus Panz. 

Luzon, Laguna, Mount Maquiling. 

Alphitobius laevigatus Fabr. 

Spec. Ins. (1781) 90; Syst. El. 1 (1801) 117; Blair, Ann. & Mag. 
Nat. Hist. VIII 13 (1914) 486, Syn. piceus Oliv. (v. Gebien, Col. 
Cat. p. 405 J). 

Genus PHAYLLIDIUS novum 

Parallelseitig, sehr flach, elliptisch, gefliigelt. Kopf unge- 
hornt, ziemlich flach, die Augen sind breit und gross, grob fazet- 
tiert, quer, ihr Abstand etwas breiter als der Querdurchmesser 
eines Auges; hinter den Augen befindet sich eine sehr feine 
Augenfurche, die Wangen sind viel schmaler als die Augen, der 
Clypeus ist nicht ausgerandet, Fuhler schlank, die Glieder zur 
Spitze verbreitert, ohne eigentliche Keule, die Mandibeln sind 
zweispitzig, scharf, ragen aber nicht vor; das Mentum ist quer 
trapezisch mit erhohter Mittellinie, die beim Mannchen einen Po- 

452 The Philippine Journal of Science 1921 

renpunkt tragt, das Endglied der Maxillarpalpen ist abgestutzt 
oval, nicht beilformig. 

Der Halsschild ist ganz quer, die Vorderwinkel treten durchaus 
nicht vor, die Basis ist gerandet und kraftig doppelbuchtig, 
die Breite ist beim Mannchen grosser als die der Decken, beim 
Weibchen so gross, bei ersterem findet sich vorn jederseits eine 
flache Schwiele. Das Schildehen ist quer. Die Fliigeldecken 
haben vollstandige Epipleuren, Oberflache mit Punktstreif en, der 
Seitenrand ist von oben nicht sichtbar, das Pygidium ist bedeckt. 

Das Prosternum ist hinter den Huften etwas verbreitert, nicht 
wie bei Phaylltts ganz parallel, das Mesosternum ist sehr tief 
und hoch V-f ormig ausgeschnitten, die sehr grossen Gelenkhohlen 
der Mittelhiiften sind seitlich offen, Hinterbrust und Abdomen 
sind vorn scharf und voUstandig gerandet, der Abdominalfort- 
satz ist ziemlich spitz, die Beine sind kurz, die Schenkel massig 
dick, ungezahnt, die Schienen gerade, aussen ausserts f ein gesagt, 
Fusse schlank. 

Die kleine in diese Gattung gehorige Art ist Phayllus aus 
SUdamerika sehr ahnlich und hat fast gleiche Kopfbildung; sie 
unterscheidet sich durch den sexuellen Dimorphismus an Hals- 
schild und Mentum, durch das tief ausgeschnittene Mesosternum 
und das schmale letzte Tasterglied. Aeussere Aehnlichkeit 
zeigt auch Epipedodema Geb. von Westafrika, hat aber sehr 
breite Epipleuren, dicke Schenkel, andere Ftihler, und seine 
Wangen sind viel breiter als die Augen. Von den bekannteren 
Gattungen durf te Sitophagus am nachsten stehen, der aber durch 
andere Taster und die BewafFnung des Kopfes beim Mannchen 
verschieden ist. 

Phayllidius dispar sp. nov. 

Einem kleinen Alphitobius diaperinus an Gestalt sehr ahnlich, 
gelbbraun, glanzend, nackt, ziemlich depress, der Rand des 
Korpers ringsum aber ziemlich steil. 

Kopf mit queren Augen, an derem Innenrand beim Mannchen 
und Weibchen ein stumpfes Winkelchen sich befindet, wodurch 
der Innenrand etwas erhoht erscheint; hinten findet sich eine 
sehr schmale Furche und ein ausserst feines Kielchen. Die 
schmalen Wangen verengen sich von den Augen an, der Vor- 
derrand ist gerade abgestutzt, die Punktierung des Kopfes ist 
fein und dicht, die Clypealsutur ist gut ausgepragt. Die Ftihler 
uberragen die Mitte des Halsschildes etwas, Glied 3 ist kaum 
langer als 4, die letzten 7 Glieder bilden eine sehr schlanke 

19, 4 Gebien: Philippine Tenebrionidse, II 453 

Keule, die vorletzten Glieder sind fast doppelt so breit wie lang, 
das letzte so lang wie breit. 

Der Halsschild ist fast doppelt so breit wie lang, die doppel- 
buchtige Basis ausserst fein und vollstandig gerandet, die Vor- 
derecken treten nicht vor, die Seiten sind in beiden Geschlechtern 
stark gerundet, etwas hinter der Mitte am breitesten, beim 
Mannchen breiter als die Fliigeldecken in der vorderen Halfte. 
Das Pronotum ist oben schwach depress, fallt aber besonders 
beim Mannchen nach den Seiten zu steil ab, und ist in diesem 
Geschlecht vom hinter den Augen jederseits stumpfbeulig 
gewolbt und dazwischen schwach eingesattelt. Die Punktierung 
ist sehr fein und massig dicht, die Hinterwinkel sind verrundet 

Die Fliigeldecken sind auf zwei Drittel ihrer Lange parallel, 
sie haben kraftige Punktstreifen, die gegen die Spitze vertieft 
sind, Streifen 1 und 2 sind an der Basis miteinander verbunden. 
Die vorn mehr order minder flachen Zwischenraume sind sehr 
fein aber deutlich punktiert. 

Das Prosternum ist iiberall deutlich punktiert, ebenso das 
Abdomen auf den ersten Segmenten. Das erstere hat vorn 
beim Mannchen eine anliegende, goldgelbe, nicht sehr dichte 
Behaarung. Die Schenkel sind massig dick, die Schienen in 
beiden Geschlechtern gerade, der Aussenrand aller Schienen ist 
sehr fein und regelmassig gekerbt, an den Hintertarsen ist 
Glied 1 so lang wie 2 und 3 zusammen. 

Lange: Weibchen, 3.7 bis 7.4 Millimeter; Mannchen, 4 bis 4.5. 

Luzon, Laguna, Los Baiios, 4 Mannchen und 3 Weibchen. 

Leptoscapha subpubescens sp. nov. 

Glanzend gelbbraun, flach, massig gestreckt, Fuhler und Beine 
hellgelb, der ganze Korper fein und wenig dicht, kurz behaart. 

Der Kopf ist flach, dicht und fein punktiert, die Augen viel 
kleiner als bei L. spissicornis, hinter ihnen befindet sich eine 
schwach abstehende Behaarung, sie sind vom kaum eingeschniirt 
und etwas f einer f azettiert als bei der madegassischen Art ; die 
Clypealsutur ist eine fein eingegrabene Linie; die Fuhler sind 
schlank und gegen das Ende deutlich kompress, die Glieder aber 
besser von einander abgesetzt als bei spissicomis, die vorletzten 
so lang wie breit, sie sind vom vierten Gliede an schwach erwei- 

Der Halsschild ist flach, an den Seiten stark gerundet, andert- 
halbmal so breit wie lang, an der Basis am breitesten, die Seiten 

181287 5 

454 The Philippine Journal of Science in\ 

im starken Bogen nach vorn verengt, die Vorderecken ragen kurz 
spitzig vor, die Mitte des ungerandeten Vorderrandes ist gerade, 
die Basis sehr fein gerandet; die Oberflache ist sehr fein, gleich- 
massig, wenig eng punktiert und ausserst fein anliegend behaart, 
nur vorn jederseits sind die Harchen aufrecht, und nahe dem 
Rande befinden sich zwei aufrechte Haare. Das Schildchen ist 
stark quer. 

Die Fliigeldecken sind vor der Mitte am breitesten, nach vorn 
wenig verengt, flach, die Seiten stark gerandet abgesetzt, die 
Schultem deutlich gewinkelt; Punktstreifen fehlen, scheinen aber 
als dunkle Linien durch. Die ganze Oberflache ist sehr fein, 
gleichmassig, wenig eng, verworren punktiert und anliegend be- 
haart, nahe den Schultern sind die Harchen deutlich aufrecht. 
Die Epipleuren sind abgekiirzt. 

Die Unterseite ist ebenfalls fein behaart, das Prostemum ist 
stumpf gekielt, zwischen den Hiiften ungef urcht, das Mentum ist 
flach, fein punktiert, das Mesosternum steigt flach an. Die 
Beine sind massig lang, die Hinterschenkel verdickt, die Tarsen 
sind dtinn, an den hinteren ist Glied 1 kaum langer als 2. 

Lange, 4 Millimeter. 

Luzon, Laguna, Mount Maquiling, 1 Exemplar. 

Ich stelle nur mit Vorbehalt diese Art in die Gattung, von 
welcher bisher nur zwei madegassische Arten bekannt sind. Mir 
liegt nur L. spissicornis vor. Die weite Entfemung der Vater- 
lander wiirde eine Trennung auch auf geringere Merkmale 
rechtf ertigen. Ich erinnere mich aber, dass die Abor-Expedition 
eine Art aus dem nordlichen Indien mitgebracht hat, leider liegt 
mir das Material augenblicklich nicht vor. Von spissicornis ist 
unsere Art sofort durch den behaarten Korper, die verworren 
punktierten Fliigeldecken, den nicht trapezischen Halsschild, die 
geringe Grosse und die gelbe Farbung zu trennen. 

Hypophloeus sulcifrons sp. no v. 

Dunn zilindrisch, von der Gestalt und ungefahren Grosse des 
H. unicolor. Glanzend schwarzbraun, die Fliigeldecken kasta- 
nienbraum, Beine, Taster und Fuhlerspitze gelblich. Der Kopf 
ist flach, die Clypealnaht sehr fein angedeutet, die Augen gross, 
rund, der Canthus den Augen vorgelagert und deutlich abgesetzt, 
nicht wie bei H, analis gegen das Auge verschwindend, der Vor- 
derkopf von den winklig abgesetzten Wangen an parallelseitig. 
Seine Seiten sind von vorn bis hinten gleichmassig aufgebogen, 
so dass innen, hart am Rande, eine lange, furchige Vertiefung 

19, 4 Gebien: Philippine Tenebrionidde, II 455 

entsteht, die aber nicht eingedriickt ist; die Mittelpartie ist sanft 
gewolbt. Die Fiihler erreichen die Mitte des Halsschildes, sie 
sind ziemlich schlank, died 3 anderthalbmal so lang wie 4, die 
folgenden quer, die vorletzten ein und zweidrittelmal so breit 
wie lang, das letzte ist schmaler als das vorletzte, der Vorderrand 
des Kopfes ist sanft ausgeschnitten, die Punktierung dicht und 
sehr deutlich. Das Mentum ist quer trapezisch, flach, sehr dicht 
punktiert, es ist aber nicht annahernd so stark quer wie bei 
unicolor, die Aussenlade der Maxillen ist sehr gross und bei 
normaler Lage der Mundteile stark sichtbar. Die Augen reichen 
unten fast bis an die Wurzel der Maxillen. 

Der Halsschild ist viel langer als breit, im grossen und ganzen 
parallel, aber eben vor der Mitte am Seitenrand sanft eingebuchtet 
und direkt vor den Hinterecken eingezogen, so dass diese selbst 
schwach aber deutlich vorspringen, die Vorderecken sind deut- 
lich, der Vorderrand gerade abgestutzt, die Basis in weitem 
Bogen nach hinten vorgezogen, voUstandig gerandet, die Punk- 
tierung gleichmassig fein und wenig dicht. Das Schildchen ist 
quer elliptisch. 

Die Seitenrandkante der Fliigeldecken hort an der Schulter- 
beule auf ; die Skulptur besteht aus feinen Reihen nicht einge- 
druckter, runder Punkte, die Reihen sind stellenweise etwas 
unregelmassig, die ganz flachen Interstitien mit einer unregel- 
massigen Reihe von fast ebenso grossen Zwischenpunkten, an 
der Spitze sind die Decken ganz verworren punktiert. Das 
Pygidium ist stark gewolbt, ohne Auszeichnung. 

Das Prosternum ist vor den Huften flach querrunzlig, zwischen 
den Huften parallel, schmal und nach hinten schwach gesenkt. 
Die Mitte des Abdomens ist sehr dicht und ziemlich grob 
punktiert und dadurch matt, das Analsegment ist ohne Eindruck, 
seitlich finden sich die gewohnlichen starken Langseindriicke. 
Die Beine sind massig lang, die Schienen gerade, die vorderen 
zur Spitze sanft verbreitert, mit spitzen Aussenendwinkeln, 
ihre Innenseite ganz sanft S-formig geschwungen, ohne Be- 

Lange, 5.5 Millimeter. 

Luzon, Laguna, Mount Maquiling (1205), 1 Weibchen, 

Die zweite Art von den Philippinen, wie es scheint, viel seltener 
als die verbreitete Hypophloeus anaUs, von dem unsere Art 
sofort durch den dunklen Vorderkorper, einfaches Analsegment, 
langen Vordefkopf mit stark aufgeworfenen Randem und 
anderen Halsschild verschieden ist. AjBhnlich ist auch H. colydir 

456 ^^^^ Philippine Journal of Science 1921 

oides Lew. von Japan, hat; aber eine einfarbig schwarzbraune 
Oberseite, ganz andere Kopfbildung und weitlaufig punktiertes 

Hypophloeus analis Geb. 

Luzon, Mount Banahao {A029) und Mount Maquiling, weiteres 

Eutochia lateralis Boh. 

Luzon, Laguna, Los Baiios (1193). 


Setenis sulcigera Boisd. Tafel 1, Fig. 4, Mannchen. 
Setenis aequatorialis Blanch. 

Mindanao, Iligan {i038). Luzon, Tayabas, Malinao. 

Setenis manillarum Fairm. Tafel 1, Fig. 5, Mannchen. 
Encyalesthus bisinuatus sp. nov. Tafel 1, Fig. 6, Mannchen. 

Glanzend kohlschwarz, ohne Spur von Metallschimmer, gross, 
robust. Der Kopf is flach, ausserordentlich fein, nur vom etwas 
grober punktiert, die Clypealsutur ist sehr tief , ihr oberer Rand 
zweibuchtig, ihr Unterrand gerade, dadurch bilden sich zwei 
mondartige Vertief ungen ; die Ausmiindungsstelle ist nicht durch 
einen feinen Ausschnitt gekennzeichnet, der Vorderrand ist ganz 
sanft ausgeschnitten ; die Augenfurche ist fein und geht, sich 
verflachend, um das Auge herum, Augenfalten fehlen. Das 
Mentum ist scharf langsgekielt und mit einzelnen langen Haaren 
besetzt, die Mandibeln sind gerade abgestutzt. Die Fiihler sind 
schlank und erreichen die Basis des Halsschildes, sie haben eine 
schlanke 5-gliedrige Keule, die vorletzten Glieder sind nur wenig 
breiter als lang ; Glied 3 ist anderthalbmal so lang als 4. 

Der Halsschild ist anderthalbmal so breit wie lang, kugelig 
gewolbt, aber oben abgeflacht, die Mitte der Lange nach ganz 
leicht angedeutet, die Sdtenrandkante ist sehr fein, die Basis 
sehr dick, in der Mitte noch starker gerandet, die Randlinie 
vorn in der Mitte undeutlich; die Punktierung ist sehr fein und 
wenig weitlaufig, die Pleuren sind glatt. 

Die Fliigeldecken sind fast zilindrisch, ihr Seitenrand ist von 
oben nicht sichtbar, sie sind tief gestreift, die Zwischenraume, 
besonders zur Spitze, kraftig gewolbt, ausserst fein punktiert; 
die Punkte der Streifen sind gleichmassig fein, regelmassig und 
bis zur Spitze deutlich. 

Die Unterseite ist nackt, das Prosternum ist hinten ganz nie- 
dergebogen, aber nicht plotzlich, der Lappen breit, zwischen 

19. 4 Gebien: Philippine Tenebrionidae, II 467 

den Hiiften ist die Brust flach vertieft, nicht deutlich gefurcht. 
Das Mesosternum ist sanft eingedriickt, das Metasternum ist 
vorn mit querer Furche versehen ; das Abdomen ist ausserst f ein 
punktiert, das Analsegment ungerandet. Die Schenkel sind 
massig stark gekeult, nackt, die Vorderscbienen etwas gekriimmt 
und innen in den letzten zwei Dritteln stark gelb bebaart, die 
Mittelscbienen dort scbwacber, die Hinterscbienen im letzten 
Drittel sebr f ein ; an den Hintertarsen ist Glied 4 so lang wie 1 
bis 3, i so lang wie 2 und 3 zusanunen. 

Lange, 17 Millimeter; Breite 6.5. 

Ein Manncben von Mindanao: Butuan. Zwei Manncben und 
drei Weibcben von Mindanao, von Herm Timm, Zoppot, erbalten. 
{Peters leg.) 

Diese Art ist von den anderen beiden pbilippiniscben sofort 
durcb die Grosse und die ganz scbwarze Farbung iiberdies aucb 
durcb die Stirnfurcben und die Beinbildung gescbieden. Nabe 
verwandt ist E, morio Geb. von Borneo, aber grosser, mit ganz 
flacben Zwiscbenraumen, kiirzeren Fublem und ganz anderer 

Encyalesthus nitidipennis Fairm. 

Mindanao, Butuan, 1 weiteres Exemplar. 
Derosphaerus rotundicoUis Cast. 

Luzon, Laguna, Los Barios {2120). 
Toxicum erythromerum sp. nov. Tafel 1, fig. 7 und la, 

Scbwarz, matt, die Scbenkel leucbtend gelbrot, verbaltnis- 
massig flacb, ziemlicb gedrungen. 

Der Kopf bat den bei den Manncben gewobnlicben, sebr tiefen, 
balbkreisformigen Eindruck, der bocb und sebr scbarfkantig 
binten begrenzt ist, die Punktierung ist ganz binten grob, in der* 
Mitte feiner, und feblt am Vorderkopf ganz, die Wangen sind 
ausserordentlicb fein und dicbt punktiert, ganz verrundet, und 
nur so breit wie die Augen. Am Vorderrand finden sicb beim 
Manncben zwei, im Winkel von etwa 80 Grad auf einander 
stebende, nackte, ziemlicb lange Horner, die am Grunde stark 
verflacbt sind und innen miteinander verbunden, ibre ausserste 
Spitze list etwas nacb binten gericbtet. Am Innenrand der 
Augen finden sicb zwei sebr lange, dunne, mit der Spitze 
gegeneinander gericbtete, und sicb fast beriibrende Homer, die 
in ibren letzten zwei Dritteln lang goldgelb bebaart sind. Die 
Fiibler baben eine starke, 4-gliedrige Keule, Glieder 6 bis 8 

458 ^^^ Philippine Journal of Science i»2i 

sind innen beim Mannchen mit einzelnen langen Haaren ver- 
sehen, das Mentum ist flach, sehr fein punktiert und nackt. 

Der Halsschild ist auflfallend flach, die Vorderecken treten weit 
vor, ihre Spitzen sind kurz verrundet rechtwinklig und etwas 
nach innen gerichtet, die Hinterwinkel scharf recht- oder gar 
etwas spitzwinklig, die Mitte ist fein angedeutet, die Punktierung 
ist ziemlich fein, nach den Seiten hin grober, aber am Seitenrand 
ganz fehlend. 

Die Fliigeldecken sind ganz flach, ihr Seitenrand ist von oben 
nicht sichtbar; die Punkte der Streifen sind fein, regelmassig, 
zur Spitze etwas deutlicher. Das Prosternum ist vorn fein, an 
den Pleuren in der Mitte sehr grob punktiert und etwas 
gerunzelt, es ist ganz flach und in einen wagerechten, spitzen, 
etwas iiberhangenden Fortsatz ausgezogen, oben deutlich doppelt 
gefurcht, das Mesosternum ist tief, senkrecht eingedriickt und 
oben mit starker Doppelfurche versehen; das Abdomen nackt, 
fein und wenig dicht punktiert, das zweite und dritte Segment 
an der Seite gerandet, das Analsegment ungerandet ; Beine ausser 
der Farbung ohne Auszeichnung. 

Lange, mit fiomern, 15 Millimeter; Breite, 5. 

Luzon, Tayabas, Malinao H022), und Mount Maquiling, 

Noch eine hiibsche Art aus der Gruppe T. flavofemoratum und 
dem T, planicolle Geb. auf den ersten Blick sehr ahnlich, aber 
kleiner und mit ganz anderer BewafFnung des Kopfes versehen, 
auch hangt das Prosternum hinten uber. Die Art ist hinter 
planicolle in meiner Tabelle ^ einzureihen. 

Toxicnin quadricome Fabr. 

Mindanao, Butuan iiOlS, JfOlQ, i020). Luzon, Tayabas, 

Anthracias elongatus Schauf. 

Palawan, Puerto Princesa. Mindanao, Butuan. Luzon, 
Tayabas, Malinao. Weiteres Material. 


Lyprops subangulatus sp. nov. 

Gross, breit, flach, oben glanzend schwarz mit ausserst kur- 
zen, anliegenden, goldgelben Harchen bekleidet; Unterseite und 
Beine schwarzbraun, Fiisse und Taster rotbraun. 

•Philip. Journ. Sci. § D 8 (1913) 400. 

19, 4 Gebien: Philippine Tenebrionidse, II 459 

Der Kopf ist grob und sehr dicht punktiert, auf dem Schei- 
tel mit blanker Stelle ; die Wangen sind so breit wie die Augen 
und ebensQ lang wie diese hinter ihnen, die Oberflache ist nahezu 
flach, der Clypeus nicht dick, die Querfurche nur angedeutet. 
Die Fuhler sind schlank, sie erreichen die Basis des Halsschil- 
des nicht ganz; Glied 3 ist nur sehr wenig langer als 4, dieses 
und die folgenden sind an Lange fast gleich, das vorletzte so 
lang wie breit, das letzte ist das grosste, etwas langer als breit. 
Das Mentum ist stark gekielt, der Kiel vom in eine vorragende, 
spitze Beule auslaufend. 

Der Halsschild ist etwa anderthalbmal so breit wie lang, 
flach gewolbt, die Seiten in der Mitte am breitesten, nach 
vorn und hinten gleichmassig verengt, die Spitze sanft ausge- 
schnitten, alle Winkel kurz verrundet, die hinteren recht- 
winklig, die Basis ist sehr fein gerandet, die Punktierung ist 
grob und an den Seiten sehr dicht; auf der Scheibe finden sich 
zwischen den Punkten zahlreiche glatte Schwielen. 

Die Fliigeldecken sind flach gewolbt, der Seitenrand sehr scharf 
glatt und in der Vorderhalfte von oben sichtbar; die Randkante 
setzt sich nach innen um die ganze Schulterbeule herum fort. 
Die Skulptur ist derjenigen von L. picinus sehr ahnlich, die 
Decken sind grob punktiert, die Zwischenraume der Punkte sind 
Uberall zu ganz kurzen Langsfaltchen ausgezogen, die beson- 
ders seitlich und vor der Spitze deutlich sind, die Bekleidung 
besteht in ganz kurzen, fast staubartigen, goldgelben Harchen 
viel ktirzer als bei picirms. 

Das Prosternum fallt hinter den Huften gerundet steil ab, 
das Mesostemum ist vom senkrecht, seine Ecken verrundet, 
etwas schwielenartig, die Propleuren sind seitlich, in der Hohe 
der Huften queruber fast glatt, das Abdomen ist sehr fein punk- 
tiert und ausserst kurz behaart, nur an den letzten beiden Seg- 
menten etwas langer. Die Beine sind diinn und schlank, an 
den Hintertarsen ist das erste Glied langer als der Rest, nur 
das vorletzte Glied mit schwammiger Sohle, an der Spitze kaum 
ausgeschnitten, es ist nicht breiter als die vorhergehenden 

Lange, 11.7 Millimeter; Breite, 4.8. 

Mindanao, Iligan, 1 Weibchen. 

Die Art sieht einem Anaedics punctatissimus von Sudamer- 
ika sehr ahnlich. Ihr nachster Verwandter ist Lyprops picinus 
Fairm. von Simalur, Sumatra, aber unsere Art ist oben schwarz, 
die Haare der Fliigeldecken sind staubartig, sehr kurz, der 
Seitenrand ist von oben in der Vorderhalfte sichtbar und 

460 ^^^ Philippine Journal of Science 1921 

geht scharfkantig um die Schulter herum, der Halsschild ist 
hinten deutlich gewinkelt, seitlich viel weniger gerundet, das 
Mentum scharfer gekielt. 

Lyprops striatopunctatus sp. no v. 

Glanzend kastanienbraun, oben lang abstehend, aber sparsam 
behaart, die Fuhler schwarz, die Kniee dunkel ; Korper ziemlich 
schlank, der Hinterkorper nach hinten erweitert. 

Der Kopf ist ziemlich flach, die Quernaht etwas eingedrtickt, 
der Vorderrand gerade abgestutzt und vereinzelt lang behaart; 
die Augen sind massig gross, ihr Abstand viel grosser als ein 
Auge im Querdurchmesser, die Wangen sind viel schmaler als 
die Augen. Die Fuhler uberragen mit den letzten beiden Glie- 
dem die Basis des Halsschildes ; Glied 3 ist so lang wie 4, beide 
langer als breit, die vorletzten sind quer, das letzte reichlich 
so lang wie 9 und 10 zusammen. 

Der Halsschild ist fast doppelt so breit wie lang, seitlich 
stark gerundet, vor der Mitte am breitesten, dahinter eingezogen 
verengt, die Hinterecken scharf, die vorderen verrundet. Das 
Pronotum ist queriiber stark gewolbt, die Basis stark gerandet, 
die Punktierung wie die des Kopfes ziemlich grob und wenig 

Die Fliigeldecken erweitern sich von der Basis bis iiber die 
Mitte, wo sie zusammen doppelt so breit sind wie der Hals- 
schild, die Schultern sind kurz verrundet, der Seitenrand ist von 
oben nicht sichtbar. Es sind starke Punktstreifen vorhanden, 
die vertieft sind, ihre Punkte stehen dicht aneinander, die ge~ 
wolbten Zwischenraume sind einzeln, ziemlich grob punktiert. 
Die Unterseite ist kaum behaart, die Beine sind schlank, das 
vorletzte Tarsenglied breiter als die vorhergehenden, und auch 
seitlich sehr deutlich behaart, an den Hintertarsen ist Glied 1 
so lang wie der Rest. 

Lange, 6 Millimeter; Breite, 2. 

Luzon, Mount Bariahao, 1 Exemplar. 

Die Art ist ganz abweichend von alien mir bekannten, durch 
die starken scharf ausgepragten Punktstreifen der Decken, 
welche, da fast alle Arten verworren punktierte Flugeldecken 
haben, zur Aufstellung einer eigenen Gattung berechtigen wur- 

Die drei Arten von den Philippinen sind weit von einander 
getrennt und lassen sich leicht so iibersehen: 

19,4 Gebien: Philippine Tenebrionidse, II 461 

1. Flugeldecken tief punktiertgestreift, einzeln lang abstehend behaart, 

letztes Fiihlerglied so lang wie die beiden vorhergehenden zusammen. 

L. striatopunctatus sp. nov. 
Flugeldecken verworren punktiert, sehr kurz behaart, letztes Fiihler- 
glied viel kurzer 2. 

2. Korper gross und sehr flach, schwarz, vorletztes Tarsenglied kaum 

breiter als die vorhergehenden, Halsschild vorn im breiten Bogen 

ausgeschnitten L. subangulatus sp. nov. 

Korper klein, gewolbt, braun, vorletztes Tarsenglied viel breiter als die 
vorhergehenden, Halsschild vorn abgestutzt L. lnzonicns Geb. 

Aediotorix petersi sp. nov. 

Von der Gestalt der andern Arten, gross, schwarzbraun, matt. 

Der Kopf ist grob, zusammenfliessend punktiert, die stark 
erhabenen Zwischenraume der Punkte bilden verworrene, stell- 
enweise kornige Runzeln, der Grund ist ausserst fein leder- 
runzlig. Die Quernaht ist leicht gebogen, deutlich eingedriickt 
und jederseits in Grubchen auslaufend. Die Wangen sind brei- 
ter als die Augen, und etwas langer, so lang wie die Schlafen, das 
Epistom ist fast parallelseitig, vorn in einen Viertelkreis ausge- 
schnitten. Die Fiihler sind ziemlich dick. Glieder 4 bis 7 sind 
deutlich quer, 8 und 9 noch starker, 10 ist so breit wie lang, 
11 ist an der langeren Unterkante fast so lang wie 9 und 10 
zusammen. Dag Mentum hat jederseits eine langliche, tiefe 
Grube, der Unterkopf hat eine sehr lange, breite, tiefe? Langs- 

Der Halsschild ist kaum breiter als lang, die Seiten sind 
kraftig gebogen und stark krenuliert, Basis und Spitze sind 
gleichbreit, der Vorderrand ist gerade abgeschnitten, die Vor- 
derecken sind deutlich, etwas flach, gedriickt, die Hinterecken 
heruntergedriickt. Die Mitte ist der Lange nach etwas ver- 
flacht, jederseits hinter der Mitte findet sich ein schrager, lang- 
licher Eindruck. Der Grund ist ausserst fein lederrunzlig, 
eigentliche grobe Punkte sind nicht mehr ausgebildet, sondern 
nur glanzende, unregelmas:sige Komer oder Runzeln, ausserdem 
sind aber, besonders vorn, deutlich getrennte, feinere, runde Kor- 
ner vorhanden. 

Jede Flugeldecke viel schmaler als der Halsschild (4.7 bis 3.5 
Millimeter). Der Skutellarkiel its schwach erhaben, die iibri- 
gen sind sehr scharf, schwach, blank, 1 lauft fast in die Spitze, 
2 ist stark verkiirzt, 2, 3, 4 enden hintereinander in einer 
gedachten Linie welche Kiel 1 parallel lauft, die Kiele (bis auf 

462 The Philippine Journal of Science 1921 

4) entspringen unmittelbar hinter der Basis. Die Zwischen- 
raume' haben 3 unregelmassige Punktreihen, von denen die 
mittlere am Vorderrand jedes Punktes eine aufrechte kurze 
Borste, aus einem Kornchen entspringend tragt, die beiden 
anderen haben nur die Kornchen, der erste Zwischenraum ist 
unregelmassig punktiert. 

Prosternum flach, am Ende mit aufgesetzter Tuberkel, Hin- 
terbrust, hinter dem Vorderrand ohne Querfurche, hinten 
jederseits neben der Mitte mit starker, runder Tuberkel. Ab- 
domen an den Seiten und an der Spitze grob punktiert. Epi- 
pleuren hinten glatt. Vorderbeine viel dicker als bei irgend 
einer der andern Arten, besonders die Schenkel, deren Unterkante 
nicht gekriimmt ist; diese hat, nahe der Mitte, einen starken, 
scharfen, aber stumpfwinkligen Zahn, nur die Vorderkante 
ist scharf ausgepragt, die hintere ganz verflacht. Mittel- und 
Hinterschenkel am Ende rundlich, aber kraftig gezahnt. Vor- 
derschiefnen sehr krumm, ziemlich dick. Mittelschienen an der 
Basis stark gekriimmt, dann verbreitert und gerade. Hin- 
terschienen, von innen gesehen, an der Basis stark gekriimmt, 
dann plotzlich verbreitert und gerade, die Verbreiterung be- 
ginnt mit einemi scharfen, breiten, abgestuzten Zahn, darunter 
befinden sich 5 lange, starke, aber nicht spitze Zahne. 

Lange', 20.8 Millimeter; Breite, 7. 

Ein Mannchen von Mindanao erhielt ich von Herrn H. Peters, 
Danzig, der sie neben andem schonen Tieren auf Mindanao 

Die grosste bekannte Art; an den dicken Vorderbeinen, deren 
Schenkel unten nicht gekriimmt sind und nahe der Mitte, nicht 
nahe der Basis, einen starken Zahn tragen, an den mit langen 
Zahnen versehenen Hinterschienen und an den zwei Schwielen 
auf der Hinterbrust sicher von den Artgenossen zu unterscheiden 

Ich dediziere diese schone Art dem Entdecker, der in der 
bereitwilligsten Weise mir sogar die Einzelstucke seiner Samm- 
lung iiberliess. 


Hemicera artactoides sp. nov. 

Sehr breit, fast halbkugelformig, wegen der Gestalt und bun- 
ten Farbung auf den ersten Blick einem Artactes sehr ahnlich, 
aber hinter der Mitte am breitesten. Die Oberseite ist sehr bunt 
gefarbt, die Unterseite dunkel blaugriin, nur das Ende des Ab- 
domens etwas bunt. 

19,4 Gebien: Philippine Tenebrionidse, II 463 

Der Kopf ist sehr kurz, der Clypeus etwa ein Drittel so breit 
wie der Abstand der Augen vorn, er ist gerade abgestutzt, die 
Naht scharf ausgepragt. Die Punktierung ist sehr fein und 
wenig dicht, die kurzen Fiihler haben eine starke 6-gliedrige 
Keule, die vorletzten Glieder sind fast anderthalbmal so breit 
wie lang, das zehnte ist an der Spitze sanft ausgeschnitten, die 
Mandibeln sind an der Spitze ganz fein ausgerandet. 

Der Halsschild ist an der Basis ungerandet, die Seiten dick 
gerandet, in der Basalhalfte fast parallel, die Vorderrandlinie 
ist in der Mitte kurz unterbrochen, die Vorderecken breit ver- 
rundet. An der Basis findet sich jederseits ein flacher, schra- 
ger Eindruck, die Punktierung ist auf glattem Grund ausserst 
fein, die Farbung sehr bunt, der Gmnd dunkel blaugriin; je- 
derseits, ferner vorn und hinten findet sich ein kupf riger, feurig 
rot umflossener Fleck. Das Schildchen ist viel langer als breit. 

Die Fliigeldecken sind viel breiter als bei irgend einer bekann- 
ten Art, weit hinter der Mitte am breitesten. Es sind Punkt- 
reihen vorhanden, die nicht yertieft sind, die Punkte sind grob 
und weitlaufig, werden aber gegen die Spitze sehr fein, die 
Punktierung der ganz flachen Zwischenraume ist kaum wahr- 
nehmbar. Die Farbung ist so bunt wie bei Hemicera splendens, 
aber nicht an die Streifen gebunden. 

Das Prosternum ist flach, ganz wagerecht, jederseits breit 
und wenig scharf gef urcht, vorn steil, aber gerundet abf allend ; 
in der Mitte des Vorderrandes findet sich eine aufgesetzte, von 
der Seite gesehen etwas zahnformig vortretende Ecke, doch 
ist nicht wie bei Hemicera splendens und anderen die Proster- 
nalplatte selbst vorn spitz gezahnt. Das scharf V-formig 
ausgeschnittene Mesosternum hat vorn scharfe Ecken, der Ab- 
dominalfortsatz ist vorn vollstandig gerandet, der Hinterleib 
selbst sehr breit und ausserst fein punktiert und lederrunzlig, 
das Analsegment hat keinen Fortsatz. Die Vordertarsen des 
Mannchens sind kraftig verbreitert. 

Lange, 6 bis 7 Millimeter; Breite, 3.9 bis 4.9. 

Mindanao, Butuan, Mannchen und Weibchen. 

Eine Art, die durch den sehr breiten Korper, die groben 
Punktreihen, bunte Farbung und die Bildung des Prosternums 
sehr ausgezeichnet ist. 

Hemicera chalcea sp. nov. 

Breit oval, hoch gewolbt, Oberseite einfarbig bronzebraun, 
stark glanzend, Unterseite und Beine schwarzblau, die Epipleu- 
ren bunt, die Wurzel der Vorderschenkel rotlich. 

464 The Philippine Journal of Science 1921 

Der Kopf ist flach, der Clypeus sanft ausgebuchtet, er ist 
etwa ein Drittel so breit wie der Abstand der Augen, die Punk- 
tierung ist sehr fein und regelmassig, die Fiihler sind sehr kurz 
und haben eine ausserordentlich breite 6-gliedrige Keule, deren 
letzte Glieder geschlossen sind. Die Ausbuchtung des zehnten 
Gliedes ist sehr breit und nimmt das letzte Glied ganz auf , das 
vorletzte Glied ist liber doppelt so. breit wie lang. 

Der Halsschild ist nicht ganz halb so lang wie breit, die Vor- 
derecken kurz verrundet stumpfwinklig, die hinteren scharf. 
Die Seiten sind fast geradlinig nach hinten verbreitert, der 
Mittellapen der Basis gerade abgestutzt. Die Punktierung ist 
fein aber deutlich, hinten etwas grober, der Grund der Ober- 
flache ist nahezu glatt, das Schildchen ist so breit wei lang. 

Die Fliigeldecken sind eben hinter der Mitte am breitesten, 
der Rand ist von oben ganz sichtbar; es sind Reihen ziemlich 
grober, r under Punkte vorhanden, zwischen je zwei Punkten 
steht ein feiner Zwischenpunkt, genau von der Grosse der fei- 
nen, weitlaufigen Punkte der Interstitien, diese sind vollkommen 
flach, kein Streifen ist vertieft, hinten wohl feiner aber nicht 

Die Prosternalplatte ist fast doppelt so lang wie breit, flach, 
jederseits gefurcht, vorn am senkrechten Absturz mit spitzig- 
em Zahnchen. Das Mesosternum ist tief V-formig ausgeschnit- 
ten und fallt scharfwinklig senkrecht ab. Der Abdominalfort- 
satz ist vollstandig gerandet, das Analsegment ohne Anhang. 
Die Punktierung ist ausserst fein. 

Lange, 7.7 Millimeter; Breite, 5. 

Luzon, Laguna, Mount Maquiling, 1 Weibchen. 

Auch diese Art ist sehr breit und ahnelt dadurch der vorigen, 
die aber ganz bunt ist und noch grobere Punkte der Fliigel- 
decken hat. Von alien bekannten Arten unterscheidet sie sich 
durch die ganz flachgedrlickte, stark verbreiterte Fuhlerkeule, 
und einfarbige Oberseite. 

Hemicera iridicolor sp. nov. 

Diese Art ist der gemeinen H, splendens Cast. & Br. von 
Java und Sumatra sehr ahnlich in Gestalt und Farbung, so 
dass es genligt, die Unterschiede hervorzuheben. Die Stirn ist 
etwas breiter, die sehr feine Augenfurche reicht etwas weiter 
nach vorn, der Grund des Halsschildes ist unter dem Mikroskop 
viel deutlicher lederrunzlig, die Streifen der Fliigeldecken sind 
durchaus nicht vertieft, nur an der aussersten Spitze einge- 
drlickt, die Punkte der Streifen sehr viel grober als die der 

19. 4 Gebien: Philippine Tenebrionidse, II 465 

Zwischenraume, wahrend sie bei splendens kaum wahrnehmbar 
sind. Die Unterseite ist wie bei der javanischen Art, also das 
Prostemum in ein nach vorn etwa& iiberragendes Zahnchen 
ausgezogen, der Abf all der Mittelbrust ist von der Seite gesehen 

Lange, 6 bis 8 Millimeter ; Breite, 3.2 bis 4.3. 

Mindanao, Butuan, 3 Exemplare. 

Aehnlich unserer Art ist auch die von Fairmaire als Eucyr- 
his gloriosus (wegen E, gloriosus Kraatz in helleri umgetaufte) 
beschriebene, aber sofort durch die matt glanzende Oberseite, 
den fast einfarbigen Vorderkorper geschieden; sie stammt von 

Die philippinischen Arten von Hemicera lassen sich wie folgt 
unterscheiden : 

1. Das Analsegment ist in einem kurzen Schwanz ausgezogen, Prosternum 

vorn ungezahnt; sehr schmale Art H. caudata Geb. 

Analsegment einfach verrundet, Prosternum vorn mit mehr oder weniger 
spitzer Ecke; breitere Arten 2. 

2. Oberseite ganz einfarbig kupferbraun, die Fiihlerkeule geschlossen, ganz 

flach H. chalcea sp. nov. 

Oberseite mehr oder minder bunt, oder mindestens auf den Flugeldecken 
mit einem griinen Langsstreifen, Fiihlerkeule locker, weniger 
flach 3. 

3. Korper fast halbkugelformig, Flugeldecken mit Reihen grober Punkte, 

sehr bunt H. artactoides sp. nov. 

Korper oval, Flugeldecken mit sehr feinen Punktreihen 4. 

4. Oberseite sehr bunt gefarbt, der Halsschild mit querer, hunter Binde. 

H. Iridicolor sp. nov. 
Oberseite einfarbig, nur die Fliigeldecken mit einem griinen Langsstrei- 
fen, zuweilen auch die Naht griinlich H. Mvittata Geb. 

Zu Hemicera gehoren auch die folgenden Arten: 

Eucyrtus pyrozonius Fairm., Ann. Soc. Ent. Belg. 40 (1896) 29, von 

Ceropria pulchra Hope, Proc. Ent. Soc. London (1842) 63; Trans. Ent. 
Soc. 4 (1845) 16. 

Hemicera zig'zaga Mars., Ann. Soc. Ent. Fr. (5) VI (1876) 111; 
Lew., Ann. & Mag. Nat. Hist. VI 13 (1894) 474, von Japan. 
Eucyrtus multicolor Fairm., Notes Leyd. Mus. 15 (1893) 40, von Borneo 
und Sumatra. Von voriger Art kaum zu unterscheiden. 

Eucyrtus auripennis sp. nov. 

Klein, sehmal, parallel, einem kleinen Tenebrio an Gestalt 
nicht unahnlich, ziemlich flach ; der Vorderkorper ist prachtvoll 
blaugriin, die Flugeldecken goldig, die Naht hinten grunlich, 
die Unterseite ist schv^arz, die Beine blaugriin, die Wurzel der 
Fiihler rot.